Note: Selected programs are available for streaming download per agreement with the original media source.
Sponsored by:
Sponsored by:
 


Subscribe to the Engineering and Computer Science Podcast (RSS) (Podcast Help)
      Audio (MP3) |  Video (MPEG-4)
  • 2008 Summer Academy for Advancing Deaf & Hard of Hearing in Computing - Speaker Series
    This program highlights deaf or hard-of-hearing professionals who spoke to University of Washington Summer Academy students about their experience obtaining advanced degrees and pursuing careers in computer science. Each speaker discusses their experiences as deaf or hard-of-hearing people in the computing industry.

    There are a variety of computing careers in industry and academia, including those that require skills in animation, games, and robotics. The Summer Academy will specifically explore animation and will provide an opportunity for students to speak with deaf and hard of hearing persons who are already working in various computing fields.



  • A Day in the Life: Kiera
    A Day in the Life: Kiera

  • A Day in the Life: Tessa
    A Day in the Life: Tessa

  • A Decidable Class of Sequentially Consistent Protocols
    Cache coherence protocols coordinate access to shared memory in multi-processor systems. A central correctness requirement for such protocols is sequential consistency: A protocol is sequentially consistent if for each protocol run, the per-processor sequences of load and store operations in that run can be interleaved to obtain a serial trace. In a serial trace, each load returns the value of the most recent store to the same memory block.

  • A Scalable, Component-driven OS
    Windows runs on a variety of hardware and with a wide range of applications. In this talk, Rob Short of Microsoft presents an overview of the challenges, tradeoffs, and technical solutions used to design and build the operating system to serve this range of masters. He discusses the details on componentization of the system, scalability, and improvements to the quality and reliability of Windows.

  • A Stateless Core Approach for Scalable Internet Services
    As the Internet evolves into a global communication infrastructure, there is a growing need to support powerful and flexible services such as traffic management and quality of service (QoS). Over the past decade, two classes of solutions have emerged: those maintaining the stateless property of the original Internet architecture, and those requiring a new stateful architecture in which routers maintain per-flow or per-connection state. Presented is a novel technique and a network architecture that bridge this long-standing gap between stateless and stateful solutions.

  • A Theory of Similarity Functions for Learning and Clustering
    Machine learning has become a highly successful discipline with applications in many different areas of computer science. A critical advance that has spurred this success has been the development of learning methods using a special type of similarity functions known as kernel functions. These methods have proven very useful in practice for dealing with many different kinds of data and they also have a solid theoretical foundation. In this University of Washington program, Maria-Florina Balcan of Carnegie Mellon University describes the theory that provides new and simpler explanations.

  • Abstract State Machines
    Computation models and commercial specification methods seem to be worlds apart. The standard computation models, e.g. Turing Machine or Random Access Machine, would not allow you even to deal with your graph (or database, etc.) directly. You have to order the graph, which is not an innocent operation; there is no known efficient method to decide whether two ordered graphs represent the same unordered graph. But the more important issue the efficiency of simulation. The ASM project started as an attempt find versatile computing devices able to simulate arbitrary computer systems directly on their natural abstraction levels. The ASM thesis asserts that abstract state machines are such versatile devices.

  • Accessible Information Technology in Education: Building Toward A Better Future
    Information technology (IT) is used in most educational settings. Students use a variety of IT tools such as email, websites, discussion boards, and courseware. They may use IT to attend school from a distance or as an adjunct to traditional classroom attendance. When these tools are accessible they can significantly reduce the effort required of individuals with disabilities and increase access to education. When they are inaccessible, they can block participation by students and faculty with disabilities. This video presents the voices of students with disabilities and experts in accessible IT as they discuss the importance of ensuring that information technology is accessible in educational settings.

  • Active Pages: Intelligent Memory for Commodity Systems
    Microprocessors and memory systems suffer from a growing gap in performance. This talk summarizes my research in Active Pages, a novel computation model that addresses this gap by partitioning computations between the processor and the memory system. An Active Page consists of a page of data and a set of associated functions that can operate upon that data.

  • Activity Recognition: Context Aware Applications
    Accurate recognition and tracking of human activities is an important goal of ubiquitous computing. The goal is to create a personal portable device that can determine a user's activity and make it available to their context-aware applications. In this talk, Gaetano Borriello and Tanzeem Choudhury present their prototype system that consists of a multi-modal wearable sensing device that communicates wirelessly to any commodity Bluetooth device and can be easily incorporated into a cell phone or a wristwatch.

  • Adaptive Algorithms: Price-Setting & Overlay Routing
    Problems of sequential decision-making under partial information commonly arise in the design of algorithms for networked systems and the applications they support. In such tasks, a decision-maker must repeatedly choose from a set of alternatives, given only partial knowledge of the past costs or benefits of these alternatives and no knowledge of the future. Classically, such problems have been modeled as multi-armed bandit problems, and they have been extensively studied and applied in a broad range of contexts including machine learning theory, economics, game theory, and the design of experiments. In this talk, Robert Kleinberg presents two such problems, motivated by applications to routing in overlay networks and pricing in e-commerce.

  • Advanced Internet Systems with Professor Dan Weld
    A summary of student's final projects in the course. Students designed their own websites and search engines and focused on searching music files on the Internet. One of these projects focused on unreliable MP3 websites with a design to make browsers avoid these sites.

  • Advances in Neural Interfaces: From Signal Processing to Optogenetics
    Could technology someday help physicians repair or replace damaged portions of the nervous system? In his talk at the University of Washington, Caleb Kemere of Stanford University/UC San Francisco says this idea is not so far-fetched. Advances in signal processing, embedded systems and photonics are helping scientists and engineers understand the human brain.

  • AEGIS Platform Architecture: Tamper-Resistant Computing
    Edward Suh briefly describes the architecture of the AEGIS secure processor and its key primitives, namely, physical nonclone-able functions and memory integrity verification. Researchers at MIT have built a tamper-resistant platform using a single-chip secure processor called AEGIS. The platform protects applications from physical attacks as well as software attacks, enabling several applications such as secure sensor networks, certified execution, and copy protection of media and software.

  • Aggregating Imprecise Data in OLAP: Principles and Algorithms
    T.S. Jayram, IBM Almaden Research Center, presents OLAP, a multi-dimensional data model in which data is analyzed across multiple hierarchical dimension attributes. Jayram considers the problem of aggregating data when there is imprecision in the hierarchy of dimension attributes. Using relevant criteria, he proposes an allocation-based mechanism to handle imprecise records. He also presents efficient algorithms of computing aggregation queries over a probabilistic database.

  • Alex St. John, CEO and co-founder, WildTangent
    Alex St. John attended the University of Alaska for a few years, but accepted a position in Boston with Hell Graphics and never graduated. He soon become the resident PostScript expert and helped make significant strides with their products. Next, he accepted a position with Harlequin Limited before trying his hand at freelancing. During his freelance consulting days he wrote several articles for prominent computer trade publications. Ziff-Davis named him one of the Most Influential People in Multimedia. Successful, but bored, he accepted an interview with Microsoft, where he was instrumental in the development of Direct X and Chromeffects. During his time at Microsoft, he became known for his wild product launches. They were wild in tone -- and also wildly effective. Soon after leaving Microsoft, he and a former colleague formed WildTangent, whose mission is to "build a richer, more exciting Internet experience for everyone - blending 3D graphics, sound, animation and interactivity."

  • Algorithms for Clustering
    Clustering problems arise in various contexts including classification, information retrieval and data mining. Roughly speaking, clustering refers to partitioning a set of objects into groups of similar objects. The objects (e.g. documents or images) are usually represented as points in some space with a distance measure and the objective is to obtain clusters of points that are close to each other. Problems of this flavor also occur in discrete location theory, where the goal is to locate a set of facilities (e.g. factories or warehouses) so as to serve a given set of clients. This talk describes several algorithmic aspects of clustering problems.

  • Algorithms for Path-Planning
    In this talk, Shuchi Chawla presents an approximation algorithm, a 3-approximation, for Orienteering. She also discusses other path-planning problems and an application of path-planning to robot navigation. Path-planning problems arise in large number of fields including operations research, scheduling, and robotics, and are mostly NP-hard.

  • Amazon.com: A Data-Driven Enterprise
    Amazon.com has grown from a small bookseller to a place where you can literally find discover and buy almost anything. As the company has grown, key lessons have been learned by leveraging customer data: where they click, what they search for, and what they buy. Special colloquia lecturer, Dennis Lee, Amazon.com, discusses the evolution of the Amazon software platform from the early days to today. He’ll highlight several areas across the company where data has been used in interesting ways that reveal very rich yet unintuitive behavior.

  • Amazon.com: Differentiating with Technology
    Jeff Bezos, CEO of Amazon.com explains the dynamic technology behind Amazon.com's store. Also find out about the company's commitment to excellent customer experience, scalability issues, real-time systems, partners and web services.

  • Amdahl's Law in the Multicore Era
    From the University of Washington, the University of Wisconsin-Madison’s Mark Hill shares his work developing a corollary to Amdahl's Law for multicore chips. This method models fixed chip resources for alternative designs that use symmetric cores, asymmetric cores or dynamic techniques that allow cores to work together on sequential execution. The results encourage multicore designers to view performance of the entire chip rather than focus on core efficiencies. This talk is part of the University of Washington Computer Science and Engineering Colloquium Series.

  • An Effective Verification Solution for Modern Microprocessors
    Over the past four decades microprocessors have come to be a vital and inseparable part of the modern world, becoming the digital brain of numerous electronic devices and gadgets that make today's lifestyle possible. However, their computational power comes at a price: the task of verifying a modern microprocessor and guaranteeing correctness of its operation is increasingly challenging, even for most established processor vendors. This talk describes a novel verification framework targeting specifically today's complex microprocessors.

  • An Introduction to Venture Capital
    Securing financing is a critical step for most startup companies in the progression from an initial idea to a fully staffed company shipping real products. For an entrepreneur starting a company, having a clear understanding of potential financing mechanisms can be just as important as knowledge of the technology and market opportunities. This program provides an overview of startup financing in general and venture capital in particular, addressing issues and questions including: What options are available for financing a startup, and what are their pros and cons? What do venture capitalists look for in a startup company? What are the expected success rates? What should entrepreneurs look for in a venture firm?

  • Analysis of a Mess: Schools, Computers, Training, and Workforce Development in the Digital Economy
    The United States has elevated computer training and literacy, workforce development, and closing the "digital divide" to the top of the national political agenda. But there are many contradictions, unresolved questions, paradoxes, and muddled thinking in the nation's current approach to these issues. Gary Chapman reviews these controversies and problems and describe some of the tasks for the computing profession.

  • Animating Human Motion
    Computer animations and virtual environments both require a source of motion for characters and objects in the environment. By using techniques to simulate humans, avatars that are responsive to the user's subtle gestures in a virtual environment are being created. Jessica Hodgins of Carnegie Mellon University discusses control algorithms that allow rigid body models to run or bicycle at a variety of speeds, bounce on a trampoline or perform a handspring vault and platform dive.

  • Applied Geometry
    From 3D surfaces in graphics to higher dimensional manifolds in mechanics, computational sciences must deal with sampled geometric data on a daily basis, leading to an increasing interest in applied geometry. This program focuses on the problem of manifold approximation and, in particular, the NP-hard problem of best approximation of 3D surfaces.

  • Apprenticeship Learning for Robotic Control
    Research into robotics is nothing new, but have we hit a point in robotics development at which we can teach robots instead of merely create them? Pieter Abbeel of Stanford University discusses apprenticeship learning techniques that have opened a door of possibilities for robotics, enabling a quadruped robot to traverse challenging terrain and a helicopter to perform difficult aerobatics.

  • Approximate Replication
    Replication is fundamental to managing data in distributed computing environments. Chris Olston describes a new framework for data replication called TRAPP (Tradeoff in Replication Precision and Performance), in which approximate replicas of remote data values are maintained in lieu of exact copies so replicas need not be refreshed with every update.

  • Approximation Algorithms for some Clustering and Classification Problems
    Clustering and classification problems arise in a wide range of application settings from clustering documents, placing centers in networks, to image processing, biometric analysis, language modeling and the categorization of hypertext documents. The applications mentioned above give rise to a number of related algorithms problems, each of which are NP-complete. Some of the general techniques, and recent developments in approximation algorithms for these problems are surveyed.

  • ARPA-E: Addressing the Sputniks of our Generation
    The report "Rising Above the Gathering Storm" proposed the creation of the Advanced Research Projects Agency - Energy. The report suggested ARPA-E to be modeled after DARPA, which was created in 1958 in response to the launch of Sputnik.

    The U.S. now faces new Sputnik-like challenges, including energy security; maintaining a technological lead; and greenhouse gas emissions and climate change. In many cases, we as a nation are lagging behind and need to change course with fierce urgency. Arun Majumdar of ARPA-E explains the agency's goal is to help catalyze this change by attracting the best minds to focus on major technical challenges and stimulating the technical and entrepreneurial community to create innovative energy technologies.



  • ASM Models using AsmL
    Yuri Gurevich, senior researcher at Redmond, WA-based Microsoft Research, leading the Foundations of Software Engineering group, explains why executable specifications will change the way software is designed, developed, tested and documented. The specification language AsmL, developed by the FSE group, makes writing ASM models practical, and allows developers to experiment with their design, validate it and enforce it.

  • Assisted Cognition
    Assisted cognition systems are proactive memory and problem solving aids that help an individual perform the tasks of day-to-day life. Henry Kautz describes the goal of the Assisted Cognition project: to create novel computer systems that will enhance the quality of life of people suffering from Alzheimer's Disease and similar cognitive disorders.

  • Automatic I/O Prefetching Hints through Speculative Execution
    Many applications, ranging from simple utilities to databases, manipulate data too large or infrequently used to be reliably found in memory caches. Due to the huge disparity between processor speeds and disk access times, these disk-bound applications tend to waste a disproportionate percentage of their execution times waiting for disk requests to complete. We can hide disk latency from these applications by taking advantage of under-utilized I/O resources to perform prefetching. To be effective, however, prefetching requires timely and accurate information about what data will be needed in the future.

  • Automatic Software Testing
    Modern software pervasively uses structurally complex data such as linked data structures. The standard approach to generating test suites for such software is tedious and error prone. This program focuses on Korat, a new technique that automates the generation of suites with structurally complex test inputs. Korat allows the developer to describe the properties of valid inputs using a familiar implementation language such as Java. Korat tools have been implemented and used in both academia and industry.

  • Automatic Tools for Building Secure Systems
    Dawn Song explains why building a secure system is a complex and error-prone process in computing. System signers and developers face many challenges. Song proposes a new automatic approach for building security protocols. Song designed and built a suite of automatic tools, Athena, containing three components: 1) APV: an Automatic Protocol Analyzer; 2) APG: an Automatic Protocol Generator; 3) ACG: an Automatic Code Generator. This toolkit enables a new automatic approach for building security protocols that is more efficient, economical, and with higher security guarantee than the current approach.

  • Automating the Design of Visualizations
    In this colloquia, Agrawala presents a new two-step approach for building automated visualization design systems. The first step is to analyze the best examples of hand-designed visualizations within a given domain. The second step is to encode these design principles as constraints within an optimization framework and algorithmically find the best visualization design.

    Agrawala describes how a two-step approach was used to develop LineDrive, a real-time system for automatically designing route maps. Agrawala describes how this project parallels the LineDrive project and will show some early results.

  • Autonomous Computing
    Pat Helland of Microsoft Research examines the notion that today's new computing model revolves around the interactions between computers that don't trust each other, which he calls autonomous computing.

  • Back to Nature for the Next Technology Revolution
    Engineering researchers such as Babak Parviz are studying nature on the nanoscale to create the next technology revolution. Imagine using DNA as a template to "grow" electronic devices, or custom designing molecules to build transistors. It could transform our future.

  • Beyond Hubs and Authorities: Web Resource Discovery and Segmentation
    After crawling and keyword indexing, the next wave that has made a significant impact on Web search is topic distillation: analyzing properties of the hyperlink graph for enhanced ranking of Web pages in response to a query. Hyperlink induced topic search (HITS) and PageRank (used in Google) are two examples.

    We discuss two enhancements to the graph selection process. First we will describe a learning system called a "focused crawler". Second we will discuss a fine-grained model for 'micro-hubs' and new algorithms based on the Minimum Description Length principle.

  • Beyond Oil: Powering the Future
    Transportation consumes 70 percent of the oil used in our country. But as worldwide demand for oil soars and supplies tighten, how will we keep transportation moving? Current and emerging technologies can quickly convert a wide range of plant matter to transportation biofuels, offering a partial solution and contributing to an increasingly diversified and "greener" energy future.

  • Beyond the Information Superhighway: Searching for the Next Policy Metaphor
    Leading policy makers, academic experts, and industry representatives convened in at The Center for Law, Commerce and Technology at the University of Washington School of Laws annual conference. Presentations focused on legal and policy questions arising from the Internet and technology areas, current developments in case law, and the influence that emerging technologies will have on industries and society. This event was sponsored by The Center for Law, Commerce and Technology; The Washington Law School Foundation; and The Washington Software Alliance.

  • Bill Gates Unplugged: On Software, Innovation, Entrepreneurship, and Giving Back
    University of Washington President Mark Emmert and the department of Computer Science and Engineering host Microsoft co-founder Bill Gates for the final stop of his six-university tour, as Gates transitions from Microsoft to the Bill and Melinda Gates Foundation.

  • Bing: Intent, Knowledge and Decision Engine
    Microsoft released its new “decision engine” in 2009. Take a look at the design decisions and technology directions behind Bing with Harry Shum, Corporate Vice President of Microsoft Corporation. The search paradigm is shifting from "search hit-or-miss model" to "Bing dialog model" where the search engine of the future must focus on improving user experience to facilitate task completion.

  • Bioinformatics: The Search for Non-Coding RNA
    One of the biggest users of scientific computing cycles in Europe is a bioinformatics application -- genome-wide searches for 'non-coding RNAs' (ncRNAs) routinely monopolize 1000 computers for an entire month. ncRNAs are functional RNA molecules that do not code for proteins. Covariance Models (CMs), statistical models based on probabilistic context-free grammars, are the leading approach to describing ncRNA families and searching for new members. In this colloquia, Larry Ruzzo describes his development of novel algorithms to make CMs faster, which allows genome databases to be scanned in days instead of years, and can greatly facilitate biological discovery.

  • Bionic and Bio-ionic Neural Interfaces
    Open your eyes to the latest developments in retinal prosthesis, which could restore vision to patients suffering from diseases such as age-related macular degeneration and retinitis pigmentosa. Luke Theogarajan of MIT discusses the two leading approaches to retinal prosthesis, a novel bio-ionic neural interface and one that is electrically based, and the hope it holds for researchers and patients alike.

  • Boolean Methods for Arithmetic Reasoning
    Detecting and correcting errors in computer systems before run-time is increasingly important in today's ubiquitous computing environment. The success of design verification and static program analysis, however, depends on how efficiently their underlying decision problems are solved in practice. Imprecise modeling often results in many false alarms and the inability to verify properties about overall system functionality and timing. Sanjit Seshia presents an approach towards tackling this problem based on new, efficient decision procedures for first-order logics involving arithmetic.

  • Borealis: Distributed Stream Processing Engines
    Recently, a new class of data management applications has emerged in areas such as sensor-based environmental monitoring, financial services, network monitoring, and military applications. These stream processing applications require low-latency processing of large-volume data streams. Because traditional database management systems are ill-suited for high-volume, low-latency stream processing, new systems, called stream processing engines (SPEs), have been developed. In this talk, Magdalena Balazinska presents the software architecture and algorithms in Borealis, one of the first distributed stream processing engines.

  • Bringing Sensing to the Masses: Infrastructure Mediated Sensing
    The use of sensing systems in the home has the potential to impact various research areas such as chronic care management, aging in place, and sustainability. But such sensing systems can also affect our daily lives. Shwetak Patel of the Georgia Institute of Technology developed what he calls infrastructure mediated sensing, or IMS. In this video, explore the possibilities of IMS, the challenges it faces and how it may enable our homes to sense our activities as well as assist in home maintenance.

  • Broadcast Encryption, Traitor Tracing, Watermarking, Dynamic Traitor Tracing, and Other Inhabitants of the Crypto Zoo
    This talk describes some recent and not so recent work on non-typical cryptographic problems having to do with the protection of intellectual property in a broadcast environment.

    It gives an overview of the strongly interrelated concepts of Broadcast Encryption (Fiat & Naor), Traitor Tracing (Chor & Fiat & Naor), Watermarking (Boneh & Shaw), Dynamic Traitor Tracing (Fiat & Tassa), and subsequent observations on these creatures. Some major open problems will be discussed as well.

  • Broadening Computer and Robotics Education and Participation for Women
    Women and other underrepresented groups represent a vast amount of untapped human resource potential needed to fuel both industry and academic research needs. Professor Andrew Williams describes a cohesive, integrated approach to increase the participation and education of women and African Americans using innovative robotics and computer curriculum and competitions. Williams provides several examples, including how the all-women Spelman College's SpelBots RoboCup Four-Legged robot soccer team, and the joint Spelman and Carnegie Mellon University NSF-sponsored project, C.A.R.E., have inspired young girls to pursue education and research in robotics and artificial intelligence.

  • Browsing Around a Digital Library
    What will it be like to work in the digital library of the future? We begin by browsing around an experimental digital library of the present, glancing at some collections and seeing how they are organized. Then we look to the future. Although present digital libraries are quite like conventional libraries, we argue that future ones will feel qualitatively different. Readers--and writers--will work in the library using a kind of context-directed browsing. This will be supported by structures derived from automatic analysis of the contents of the library--not just the catalog, or abstracts, but the full text of the books and journals--using new techniques of data mining.

  • Bud Albers, CTO, Getty Images
    Bud Albers is responsible for Getty Images’ worldwide technical operations and provides strategic technology counsel for the Getty Images executive team. His areas of responsibility include technology development, implementation and support. Bud joined Getty Images from Monsanto Corporation, where he served as Chief Technologist and head of enterprise application development. Prior to that, he was Chief Technologist for Ameritech’s e-commerce division, and held various senior operational and technology management roles at NxTrend, a leading developer of enterprise software solutions.

  • Build the Future: UW Computer Science & Engineering
    An overview of the UW Department of Computer Science & Engineering, featuring student projects and interviews with regional technology and education leaders.

  • Building Secure Systems from Buggy Code with Information Flow Control
    The intensity of today’s computer security resembles an arms race: the bad guys constantly find new ways to break in and being safe requires ceaseless efforts to stay one step ahead to cut off avenues of attack. But this strategy is risky and expensive. Nikolai Zeldovich of Stanford University examines the use of information flow control to build secure systems out of buggy code and ways to reduce or even eliminate security vulnerabilities.

  • Building the Future at the University of Washington
    The Paul G. Allen Center for Computer Science & Engineering is a unique laboratory-intensive research facility on the University of Washington campus. The center’s architectural design fosters faculty research, creative student design projects and partnerships with computing industry leaders. Touring this state-of-the-art facility emphasizes the collaborative processes and innovative problem-solving that has established the CSE among the top ten computer science programs in the nation. The building design and construction was funded through generous donations by Paul G. Allen, Bill & Melinda Gates Foundation, and the Microsoft Corporation.

  • Calculi for Access Control
    Access control is central to security in computer systems. Over the years, there have been many efforts to explain and to improve access control, sometimes with logical ideas and tools. Martin Abadi, Microsoft Research/UC Santa Cruz, reviews some of that work and its applications. He also explores a new approach based on type systems, specifically on a type system for tracking dependencies.

  • Capturing Life Experiences: Automated Video Capture
    For the past ten years, computer science engineers and faculty at Georgia Institute of Technology have been motivated to achieve some fraction of the potential that Mark Weiser envisioned for ubiquitous computing. Taking an applications-driven approach, Gregory Abowd describes their research focus on augmenting spaces, including the classroom, office, home and body, providing living laboratories of ubiquitous computing research. One common theme in their investigations is the capture of live experiences for later access.

  • Chaos in Computer Performance
    Are computers inherently chaotic? New research from the University of Colorado at Boulder suggests that computer software has grown so complex as to defy traditional tools of analysis. Join Elizabeth Bradley, professor in the department of Electrical, Computer, and Energy Engineering at the University of Colorado, for a new, dynamic and non-linear approach to computer design. This video is part of the University of Washington Computer Science and Engineering Colloquium Series.

  • Chinasite.com
    ChinaSite.com is an internet portal business, the web's gateway to China.

  • Clustering Gene Expression Data
    This program presents a brief introduction to microarray technology for those unfamiliar with it, and outline recent work by Larry Ruzzo, students and other collaborators on cluster analysis of microarray data.

  • Coarse-to-Fine Natural Language Processing
    State-of-the-art NLP models are anything but compact. Syntactic parsers have huge grammars, machine translation systems have huge transfer tables, and so on across a range of tasks. With such complexity come two challenges. First, how can we learn highly complex models? Second, how can we efficiently infer optimal structures within them? Syntactic parsing, acoustic modeling for speech recognition, and machine translation issues are all discussed in this talk.

  • Cognitive Developmental Robotics: An Approach To Understand Ourselves And To Design Robots Like Us
    Cognitive Developmental Robotics (CDR) aims to provide new understanding of how human higher cognitive functions develop by means of a synthetic approach that developmentally constructs cognitive functions. From the University of Washington’s Computer Science and Engineering Distinguished Lecturer Series, Minoru Asada of Osaka University shares how the core idea of CDR is "physical embodiment'' that enables information structuring through interactions with the environment, including other agents. The idea is shaped based on the hypothesized development model of human cognitive functions from body representation to social behavior. Along with the model, Asada will introduce studies of CDR and related works in the talk, and discuss the model and future issues.

  • Collaborative Dynabooks
    A dynabook is a platform for learning through creation and exploration of multimedia. Thirty years of cognitive and learning sciences research suggests that the dynabook vision holds promise for improving education. It will work most powerfully, however, in a collaborative context. Collaborative dynabooks offer a new way of structuring how and what is taught in computer science labs and classes.

  • Collaborative Systems
    As a result of the ubiquity of computer networks, computer systems increasingly participate in complex, distributed communities of people and systems, rather than operating as solitary devices employed by a single person. This major shift in the way we use computers has led to a significant challenge for computer science: to determine ways to construct computer systems that are able to act effectively as collaborative team members. This program features one model of collaborative planning, SharedPlans, and describes other efforts to develop collaborative planning agents and systems for human-computer communication.

  • Communications for Mobile People
    People are the outsiders in the current communications revolution. Computer hosts, pager terminals, and telephones are addressable entities throughout the Internet and telephony systems. Human beings, however, still need application-specific tricks to be identified, like email addresses, telephone numbers, and ICQ IDs. The key challenge today is to find people and communicate with them personally, as opposed to communicating merely with their possibly inaccessible machines.

    The main goal of the Mobile People and IdentiScape projects is to put the person, rather than the devices that the person uses, at the endpoints of a communication session.

  • Community Systems: The World Online
    The Web is about you and me. Until now, for the most part, it has denoted a corpus of information that we put online sometime in the past, and the most celebrated Web application is keyword search over this corpus. Sites such as del.icio.us, flickr, MySpace, Slashdot, Wikipedia, Yahoo! Answers, and YouTube, which are driven by user-generated content, are forcing us to rethink the Web -- it is no longer just a static repository of content; it is a medium that connects us to each other. What are the ramifications of this fundamental shift? What are the new challenges in supporting and amplifying this shift?

  • Compiler-Directed Synthesis of Hardware Accelerators
    An ever larger variety of embedded ASICs is being designed and deployed to satisfy the explosively growing market for new electronic devices. Many of these devices handle demanding multi-media computations. In many such ASICs, specialized nonprogrammable hardware accelerators (NPAs) are used for parts of the application that would run too slowly if implemented in software on a programmable processor. Rapid, low-cost design, low production cost, and high performance are all important in NPA design. In order to reduce the design time and design cost, automated design of NPAs from high-level specifications is becoming increasingly important. A principal goal of the HP Labs Program-In-Chip-Out (PICO) project is to automate the design of NPAs.

  • Computational Biology: Comparing Vertebrate Genomes
    There are currently 17 vertebrate genomes, ranging from primates to fishes, for which we know nearly their entire DNA sequences, and this number will increase rapidly. Comparing these genome sequences has emerged as one of the most important areas of computational biology. One way to predict functional portions of the human genome, for example, is to search among related genomes for sequences that appear to be remarkably similar due to selective pressure. Martin Tompa discusses and demonstrates some of the methods and tools for such an approach, as well as some of the challenges and unsolved problems.

  • Computational Biology: Genomics
    One of the current major challenges in computational biology is to extend our knowledge to very large sets of genes and proteins. Michal Linial presents a technique that automatically clusters protein sequences, resulting in a classification of the input proteins into a hierarchy of clusters of varying degrees of granularity.

  • Computational Discovery of Genetic Regulatory Networks
    With the draft sequences of many genomes in hand, the next frontier is understanding genetic regulatory networks that control cellular behavior and development. In our approach to discovering genetic regulatory networks we first learn about modules of genes and their interactions, and then refine this understanding into a model that describes the precise interaction of individual genes.

  • Computational History in Action: Gutenberg’s Printing Process
    This colloquia outlines the computational research methods, including multi-resolution clustering, image understanding, and non-rigid registration, that were used to rediscover the revolutionary technologies that created the first typographic print-press books. Using data analysis techniques based on high-resolution digital photography of surviving print press copies, research by Blaise Aguera y Arcas has shown that Gutenberg invented two different transitional printing technologies, both of which had apparently been forgotten by 1500. The first invention resembles a midway-point between printing and scribal practices. The second invention, also unconventional, bears a striking resemblance to Linotype, the 19th century technology widely held to have brought printing into the modern era.

  • Computational Textiles and the Democratization of Ubiquitous Computing
    The blossoming research field of e-textiles integrates computation with fabric. E-textile researchers weave, solder and sew electronics into cloth to build soft, flexible and washable computational devices. E-textiles is a young discipline, and developments in the field have been relegated almost exclusively to research labs in industry and academia. In this University of Washington program, learn about advancements that make e-textiles accessible to new audiences that are helping to democratize ubiquitous computing and integrate electronic hardware with cloth.

  • Computer Architecture
    Understanding program behavior is at the foundation of computer architecture and program optimization. Sherwood describes a new technique, Basic Block Distribution Analysis, as a means of summarizing, visualizing, and exploiting the time varying behavior of programs. Many programs execute as a set of phases. Using techniques from machine learning, he demonstrates how the phases can be found automatically, at either profile or run-time. This information can then be used to reduce simulation time and help guide expensive run-time optimizations.

  • Computer Graphics: Communications Media
    David Salesin discusses how the real market for computers lies in their vast potential as a communication medium. Drawing upon examples that range from computer-generated illustration and virtual cinematography to adaptive document layout and animated presentations, Salesin discusses some of the research challenges he sees in harnessing the power of the computer to create more powerful communications media.

  • Computer Science & Engineering Distinguished Lecturer Series: Closing the Innovation Gap
    Innovation drives economic growth, our quality of life and is the only hope of addressing the major challenges we face. But America, a cornerstone of innovation throughout the world, has become increasingly short-sighted. By taking innovation for granted we threaten not only our own strength, but the overall global economy. Judy Estrin, technology and business pioneer and author of the new book Closing the Innovation Gap, will talk about how it is essential to reignite sustainable innovation in business, education and government and what is required of business and national leaders to revive organizational, national and global Innovation Ecosystems.

  • Computer Science and Engineering Distinguished Lecturer Series and Dean Lytle Electrical Engineering Endowed Lecture: From Cell Phones to Smart Phones to Smart Books - An Exciting Journey
    More than one billion cellular devices are now shipped each year to more than four billion subscribers worldwide. More than half will soon support wide-area broadband access to the Internet with devices that are increasingly more powerful, more compact and less expensive than their predecessors. Irwin Jacobs, co-founder of Qualcomm, will touch on the history of the wireless telecommunications research and development company, and explore further developments in wireless technology, devices and applications.

  • Computer Science and Engineering Distinguished Lecturer Series: Cooking in Silico: Understanding Heat Transfer in the Modern Kitchen
    During the last decade, haute cuisine has undergone a scientific revolution. Leading chefs have taken an interest in the science of cooking and in scientific tools found more commonly in research laboratories. Centrifuges, freeze dryers, digitally-controlled water baths and liquid-nitrogen filled Dewar flasks are just a few examples of technologies that have transformed the modernist kitchen. The computer remains an underutilized tool for exploring the hows and whys of cooking. In this talk at the University of Washington, Nathan Myhrvold and Chris Young of Intellectual Ventures show how computationally intense heat-transfer calculations can reveal the subtle factors that influence the success or failure of a cook's efforts in the kitchen. Explore the virtues of computational cooking, and watch novel techniques and creations made possible when science informs the culinary arts.

  • Computer Science and Engineering Distinguished Lecturer Series: Interactive Machine Learning
    Machine learning offers the promise of assisting users to create new tools simply by demonstrating the desired outcome. The tool "Image Processing with Crayons" was built for creating classifiers for image-based problems. The tool empowers a much larger class of people who can create image-based interactive techniques. However, observations of people using Crayons have pointed out several challenges in the way machine learning algorithms are designed. People do not behave in statistically uniform distributions and more importantly their interaction with the learning algorithm distorts their behavior in specific ways. The lessons we have learned will be discussed along with new directions for machine learning algorithms that might learn faster in the face of user behavior.

  • Computer Science and Engineering Distinguished Lecturer Series: Rethinking Computing
    Craig Mundie, chief research and strategy officer of Microsoft, presents “Rethinking Computing,” a look at how software and information technology can help solve the most pressing global challenges we face today. Part of UW’s Computer Science and Engineering’s Distinguished Lecture Series, Mundie demonstrates a number of current and future-looking technologies that show how computer science is changing scientific exploration and discovery in exciting ways. He also discusses the role of new science in solving the global energy crisis, and answer questions from the audience.

  • Computer Science and Engineering Distinguished Lecturer Series: Return to the Final Frontier
    Spaceflight participant and former Microsoft software developer Charles Simonyi compares his first and second spaceflights as part of the University of Washington Computer Science and Engineering Distinguished Lecturer Series. He’s observed encouraging maturity in manned space technologies, as well as valuable lessons from his experiences in space. Simonyi also shares video of his return from orbit in a Soyuz capsule.

  • Computer Science and Engineering Distinguished Lecturer Series: Robotic Cars: Challenges and Perspectives
    As the US automotive industry is at the brink of collapse, we now face a unique opportunity to "begin more intelligently," as Henry Ford once said. Today's automobile industry is wasteful along many dimensions, such as energy consumption, resource utilization, human comfort, and safety. Sebastian Thrun from Stanford University will talk about innovative ideas for "Car 2.0," which rely heavily on computer science. He will specifically address the topic of robotic cars, discuss his experiences with the DARPA Challenges, and highlight ongoing research on smarter, safer, and more efficient transportation.

  • Computer Science and Engineering Distinguished Lecturer Series: The Web the Way You Want It
    The decentralized architecture of the web was designed from the outset to create an environment where content producers and consumers could come together without the need for everyone to use the same server and client. To participate in the web revolution, users only needed to subscribe to the basic architecture of a web of content delivered via HTTP and addressable via URLs. Given this architecture, specialized browsers have always existed to a greater or lesser degree alongside mainstream web browsers. This talk will highlight specialized browsers in the context of accessibility; for use in mobile environments, or for use by persons with specific needs. As we evolve from the purely presentational web to a more data-oriented web, such specialized tools become center-stage with respect to providing optimal information access to the end-user. The talk will conclude with a brief overview of where such web technologies are headed and what this means to the future of making web content accessible to all.

  • Computer Science and Engineering Distinguished Lecturer Series: Why are Graphics Systems So Fast?
    Over the last decade, graphics hardware has become a key component of mobile and personal computers. Most programmers understand CPUs well, but have a limited understanding of GPUs (Graphics Processing Units). In this video from the University of Washington’s Computer Science and Engineering Department, Pat Hanrahan of Stanford University explains the architectures of different GPUs built by AMD, NVIDIA and Intel (the new Larrabee processor). The innovative combination of processor design and programming model makes for fast graphics systems.

  • Computer Science and Engineering Distinguished Lecturer: Declarative Networking: "What" is Next
    Declarative languages allow programmers to say what they want, without worrying over the details of how to achieve it. These kinds of languages revolutionized data management decades ago but have had limited success in other aspects of computing. The story seems to be changing in recent years, however. One new chapter is work that Joe Hellerstein and his colleagues have been pursuing on the design and implementation of declarative languages and runtime systems for network protocol specification. Distributed Systems and Networking appear to be surprisingly natural domains for declarative specifications and they are ripe for a new programming methodology.

  • Computer Science and Engineering Distinguished Lecturer: Why The Algorithm Might Soon Be The Only Game in Town
    Are our brains cuckoo clocks? Do fish swarm in an algorithmic manner? In this extremely engaging discussion, Bernard Chazelle of Princeton University utilizes such examples, as well as entertaining political and historical references, to show the power of algorithms and the effects of the fast approaching era of algorithms.

  • Computer Science and Engineering Industrial Affiliates Program: The Changing Face of Venture Capital
    Learn more about the Industrial Affiliates Program, which aims to support the mutual needs of business, industry and academia in computer research, development and education. The program offers business and industry partners opportunities to influence computing research and education, the chance to participate in long-range technical assessments of problems and directions in the field as well as a link to prospective employees. Students are introduced to industrial needs and available internships.

    The Industrial Affiliates Program offers a constructive relationship between industry and universities – a relationship that stimulates developments in both segments of our society.



  • Computer Science Education
    Stuart Reges discusses four nifty assignments,' such as one on object-oriented programming, presented at the annual symposium for SIGCSE (Special Interest Group on Computer Science Education(.

  • Computer Science Participation
    Despite decades of intervention, women and minorities remain significantly underrepresented in computer science and engineering at all levels of the academic pipeline. In this CSE Distinguished Lecture, Professor Jan Cuny, University of Oregon, reviews a number of existing efforts at the national level and calls for broad new alliances to solve this problem.

  • Computer Science Programming Languages
    A core goal of computer science is to develop methods for building reliable and maintainable programs. Robert Harper reviews some exciting new developments in language research, such as the emergence of type theory—the study of type systems—as the grand unified theory of programming languages, encompassing both language design and implementation.

  • Computer Science Research for Global Development
    On the same planet where there are 1.4 billion Internet users, a far less fortunate 1.4 billion people survive below the World Bank’s definition of the poverty line. The same technology that has transformed our lives - the lives of the wealthiest people on the planet - also remains out of reach and irrelevant for the poorest. This talk introduces the Technology for Emerging Markets group at Microsoft Research India, where an interdisciplinary team of researchers explores solutions in the context of agriculture, education, healthcare, microfinance, and other domains of development.

  • Computer Science: Past, Present and Future
    Computing research has made remarkable advances, but there's much more to be accomplished. The next ten years of advances should be even more significant, and even more interesting, than the past ten. The National Science Foundation has created the Computing Community Consortium to engage computing researchers in an ongoing process of visioning -- of imagining what we might contribute to the world, in terms that we and the world might both appreciate. This process is just beginning. In this program, Ed Lazowska reviews the progress the field has made and presents a number of "grand challenge" problems we should be prepared to tackle in the coming decade.

  • Computer Science: Still Crazy After All These Years
    Computer science is transforming all aspects of our lives, and driving our nation's economy. In the face of this, it is startling to recognize that the real computing and communications revolution still lies ahead. This talk describes recent progress, future opportunities and the special role of research universities in the field of computer science.

  • Computer Technology: Solving Public Health Challenges
    The widening economic gap between countries is paralleled by disparities in health outcomes. Computer technology has the potential to play an important role in the efforts to improve healthcare in poor settings. The useful application areas includes electronic medical record systems, decision support, healthcare education, telemedicine, data gathering, and a wide range of communication systems. In this talk, research scientist Neal Lesh provides an overview of international health issues and outlines several potential computer technology applications for communities with severe resource limitations.

  • Computers, Cultures and Constructions: Locating the Learner in the Con/Text of Digital Video Cases
    Innovative uses of video, audio, text, and graphics make it possible for the learner to engage in knowledge creation activities. In this presentation, Ricki Goldman-Segall will show examples and discuss how she has used rich media in her work with children.

  • Configuration, Customization and Appropriation: Integrating Technology and Practice in the Placeless Documents System
    A key feature in the adoption of a software system is the extent to which it can be appropriated into everyday working practice. Appropriation is a feature of both the social setting and, critically, the software design. We have been exploring these issues in the design of the infrastructure and applications for Placeless Documents. Placeless Documents is a fluid and flexible document management infrastructure. Rather than the fixed hierarchies that characterize most document storage systems, Placeless provides an extensible, composable, associative storage model that accommodates multiple users, multiple tasks and multiple application needs simultaneously.

  • Connecting the Pacific NW
    NSF Networking Initiatives
    NSF networking research and infrastructure programs have been the stimulus for the availability of advanced Internet technology to and within the Pacific Northwest, with the University of Washington playing the leadership role.

    The Wellspring of Discovery
    This 50th anniversary symposium provides an opportunity to reflect on the agency's role as a wellspring for discoveries. NSF's work has nourished discovery as well as our nation's economy; the challenge now is to sustain the exploration and to convey the "awed wonder" of science.



  • Constructing Code: Expanders
    Expanders, sparse yet highly connected graphs with additional pseudorandom properties, have been the main combinatorial tool in the explicit constructions of error-correcting codes with fast encoding and decoding algorithms. Recently, these constructions were improved to the point where the rate (a measure of efficiency of the code) vs. error-correction tradeoff is near optimal. Venkatesan Guruswami presents an overview and recent work which uses expanders and related combinatorial tools to construct codes with linear-time list-decoding algorithms. Based on joint works with Piotr Indyk.

  • Content-Preserving Warps for Video Stabilization and Wide-Angle Imaging
    Content-preserving warps allow video editors to deform images while preserving the shape and appearance of salient image content. Aseem Agarwala of Adobe describes how variants of this method can be applied in surprising ways. Content-preserving warps can improve video stabilization, transforming shaky hand-held camera work and into smooth motion video. Examine a challenging case: modifying a video shot while walking through a 3D scene so that it resembles the "tracking shots" used in films, where rail-mounted cameras are carefully pushed along ideal paths. Another problem is minimizing perceptual distortion in wide-angle images. Large field-of-view images are easy to capture with wide-angle lenses or by stitching panoramas, but the images typically look distorted; straight lines in the scene may be curved, and objects stretched or squashed. The goal is to project the visual content defined across a large field-of-view into a distortion-free image.

  • Critical Infrastructure Protection
    In this distinguished lecturer colloquia, Scott Charney details the history and current state of cybercrime. This special lecture includes a discussion of Microsoft's security initiatives and identifies the challenges that governments, industry, and the public face as they attempt to prevent and respond to computer abuse.

  • Crosslinguistic Resources: Scalable Precision Grammars
    Precision grammars represent knowledge engineering for natural language parsing (assigning syntactic and semantic structure to sentences) and generation (assigning surface strings to semantic representations). Recent advances in parsing and generation technology make it feasible to use precision grammars in natural language processing tasks demanding real&345;time response from the parsing system. In this talk, Emily Bender presents the LinGO Grammar Matrix, a linguistic knowledge resource designed to allow for rapid prototyping of scalable precision grammars of any human language.

  • CSE477 Capstone Design, Spring 2008- Technology for Low-Income Regions
    Program Description: Capstone courses give students experience solving a substantial problem using concepts that span several topic areas in Computer Science and/or Computer Engineering. Students must work together in teams to define the problem, develop a solution plan, produce and demonstrate an artifact that solves the problem, and present their work using written and oral reports. Cross-disciplinary projects that require interaction with other departments are encouraged.

    This year-long capstone design course looks at problems in health care, agriculture, transportation, and education that arise in the developing world and rural regions of the developed world.



  • Cybersecurity: The First Pacific Rim Regional Collegiate Cyber Defense Competition
    The First Pacific Rim Regional Collegiate Cyber Defense Competition—or "Geeks Under Attack!"— tested students’ ability to protect enterprise network and business information systems. Hosted by Microsoft and organized by Barbara Endicott-Popovsky, Director of the Center for Information Assurance and Cybersecurity at the University of Washington Information School, PRCCDC offered a challenging proving ground for nine teams of rising cybersecurity professionals. The documentary was funded by a generous gift from Microsoft.

  • Cyclone: Programming-Language Technology for Reliable Software
    Building large, reliable software systems is one of the great intellectual challenges of our age. UW CSE researcher, Dan Grossman, describes the ongoing efforts to use programming languages and their implementations to make reliable software easier to write. The talk includes a description of Cyclone, a type-safe programming language at the C level of abstraction.

  • Cyclone: Safe Programming
    Memory safety and type safety are invaluable features for constructing robust software, but most safe languages are at a high level of abstraction. This is one reason C remains the de facto standard for writing systems software or extending legacy systems already written in C. Dan Grossman explains how the Cyclone Project aims to bring safety to C-style programming without sacrificing the programmer control necessary for low-level software.

  • Data Mining
    Usama Fayyad presents a paradoxical technical and business dilemma: an over-abundance of data with limited ability to utilize or mine the vital information. Data mining—a technology few organizations have been successful in managing—is discussed on several levels: algorithms for scaling to large databases, aspects and challenges for fitting data mining with database systems, and finally the challenges of how to make the technology really work for e-business.

  • Data Structures & Algorithms
    Turing Award Winner, Robert Tarjan, reviews several problems that he has explored in the in the realm of data structures and algorithms. The emphasis is on problems still in need of a solution.

  • Data-Driven Texture and Motion
    Efros presents work in two different domains, describing algorithms for visual texture synthesis (e.g. grass, pebbles) and human motion from video. The method can be used for both action recognition and motion transfer.

  • Dataflow Architectures
    Several recent architectural proposals, including TRIPS and WaveScalar, advocate dataflow-like architectures as a good match for the execution substrates projected in the next decade and beyond. In this program, Doug Burger discusses the future of architecture research, describing the TRIPS prototype architecture and a distributed selective re-execution protocol. His special focus is describing a scheme which provides a unified underlying framework to aid recovery from mis-speculation and soft errors.

  • Debugging Concurrent Software by Context-Bounded Analysis
    Designing concurrent programs is difficult. The nondeterministic interaction between concurrently executing threads of a concurrent program results in programming errors that are notoriously difficult to reproduce and fix. Shaz Qadeer of Microsoft presents a new static analysis technique based on model checking for automatically finding errors in concurrent software. An execution of a concurrent program is a sequence of contexts, where each context consists of transitions performed by a single thread. KISS (Keep It Simple and Sequential), is a technique for transforming a concurrent program P into a sequential program Q that encodes all executions of P with a small number of contexts. KISS is used to convert SDV (Static Driver Verifier), a model checker for sequential C programs, into a model checker for concurrent C programs. The combination of KISS and SDV has been used to find a number of subtle concurrency errors in Windows device drivers.

  • Deduplication Storage System
    This talk gives an overview of the latest disruptive technology in the storage systems industry called deduplication storage system. Disk-based deduplication storage has emerged as the new-generation storage system for enterprise data protection to replace tape libraries.

  • Design Methods: Distributed Systems
    Heterogeneity and distribution are becoming key properties of both high-performance integrated circuits and real-time embedded systems. For instance, the embedded electronics of a modern car includes a heterogeneous mix of components (ECUs, sensors, actuators) and networks (CAN, FlexRay, MOST) that constitutes a distributed architecture. The key to addressing distribution challenges is the development of methodologies based on formal methods that enable modularity, flexibility, and reusability in system design. By working at the system level, UC-Berkeley researcher, Luca Carloni, illustrates how these methodologies can take advantage of the commonalities that exist between IC design and embedded software programming.

  • Designing Appropriate Computing Technologies for the Rural Developing World
    Tapan Parikh describes his experiences developing CAM - a toolkit for mobile phone data collection - in the rural developing world. Designing technologies for an unfamiliar context requires understanding the needs and capabilities of potential users. Drawing from the results of an extended participatory design study conducted with microfinance group members in rural India (many of whom are semi-literate or illiterate), he outlines a set of user interface design guidelines for accessibility to such users. The results of this study are used to motivate the design of the CAM toolkit, which includes support for paper-based interaction; multimedia input and output; and disconnected operation. Parikh discusses possible topics for future work and his long-term research vision.

  • Designing for Fluent Interaction
    As the world of computing moves away from individual interactive devices towards ubiquitous computer augmented environments, new issues of usability arise and others take on increased importance. In this Distinguished Lecture colloquia, Terry Winograd outlines what has been learned from these 'off the desktop' computing experiences, and discusses how to extend the theories to cope with some of the new challenges and opportunities in Human Computer Interaction.

  • Designing User Interfaces
    We are now entering the era of pervasive computing, where people will access information and services anywhere, anytime, and from a wide variety of devices. James Landay (UC Berkeley) talks about designing tools that support the best practices of user-centered design.

  • Digital Natives: Impacts on Management and Education
    How are the work styles and learning styles of digital natives different from their counterparts? Are they more competitive? How are social networks influencing the way they live and think? Join Michael Eisenberg, dean emeritus of the iSchool at the University of Washington, and a panel of experts as they discuss these topics.

  • Digital Photography: Bhutan Expedition
    Hawley describes his cutting-edge use of digital camera equipment during three MIT-led photographically-intensive expeditions in the Himalayan paradise of Bhutan. He explores print processing and web archiving using the metadata encodable in digital images and provides examples of how digital still cameras now exceed film in both quality and creative latitude.

  • Digital Simplicity through Activity-Based Computing
    Recent advances in small, inexpensive sensors, low-power processing, machine learning and mobile-user interfaces have enabled applications that use on-body sensing to infer people's activities throughout everyday life. The Digital Simplicity project brings this technology into our everyday lives and simplifies our high-level, long-lived activities. Target activities include supporting the elderly’s health and independence, motivating individuals to get fit and reducing a family’s environmental footprint. Explore these applications as well as the underlying sensing, inference and design tools required to bring Digital Simplicity to everyday life.

  • Directions and Challenges in Integrated Circuit Scaling
    The integrated circuit industry has enjoyed tremendous success over the past 30+ years due to the excellent scaling properties of MOS transistors. As MOSFETs have scaled, consistent improvements in density, performance and power have been realized. However, we are fast approaching practical and fundamental limits to scaling that will require new strategies to enable the industry to continue its historic rate of improvement. These strategies must include new device structures, new materials and new circuit design techniques.

  • Discrete Global Minimization Algorithms
    Steven Gortler demonstrates how global optima can be found as the limit of a set of purely combinatorial problems. These combinatorial problems can be thought of higher dimensional analogues of the 'shortest path in planar graph' problems. He explains how these combinatorial problems can be solved in polynomial time by reducing them to instances of MIN-CUT.

  • Discrete Mathematics: Expanders Graphs & Eigenvalues
    Expander graphs are among the most interesting objects of study in modern discrete mathematics. They are useful for a broad spectrum of applications in computer science, from the design of good routing networks to de-randomization. The study of expander graphs has brought into discrete mathematics and theoretical computer science a variety of new, powerful mathematical tools. In this talk, Nathan Linial defines the concept of an expander graph, and illustrates one application in de-randomization. Linial also explains the relationship between expansion, which is a combinatorial concept to spectral gap - a linear algebraic parameter that is easy to compute.

  • Display System Performance
    Once upon a time graphics system performance was expressed in terms of simple numbers: polygons per second, pixels per second, and frames per second. All of the numbers were low and many of us worked hard to make them better. Today the numbers are much better, but mean much less. As memory and computational costs have declined the amount of content that can be pushed through display pipelines has increased dramatically and continues to increase. This raises the question of limits. How much performance is enough?

  • Distinguished Lecturer Series: David Ditzel - A 25 Year Perspective on Binary Translation: What Worked, What Didn't
    Binary Translation is a technique that continues to grow in acceptance. It has been used for moving customer applications across systems which would otherwise be binary incompatible. Early examples include Hunter Systems XDOS and Digital Equipment's FX132, to Transitive Technologies current use by Apple Computer to help move customer applications from PowerPC to Intel processors. Binary translation techniques will become increasingly common in the future as a way to introduce new techniques in computer architecture while proving backwards binary compatibility.

  • Distinguished Lecturer Series: Jeff Dean - Research Challenges Inspired by Large-Scale Computing at Google
    This lecture gives some background information on Google's existing hardware and software infrastructure and will examine what works well and what does not, and some areas where interesting unsolved research problems are highlighted. The problems span a wide range of topics, including processor design, distributed systems, machine learning, information retrieval, text processing and many other areas. This talk is meant to cover a sampling of interesting problems/areas, not a comprehensive treatise.

  • Distributed Router Fabrics
    Today's IP routers make incredible performance demands of their switch fabric. The router fabric is responsible for moving packets between tens to hundreds of high-bandwidth ports without any disruptions, all while meeting service guarantees. As these demands continue to increase, fabrics must become highly distributed while allowing for a flexible topology so they can be economically packaged. Towles describes one solution to implement the switching functionality of an IP router using an interconnection network --- a distributed router fabric. Brian Towles is currently a Ph.D. candidate in electrical engineering at Stanford University and his thesis research focuses on distributed router fabrics. He was also involved with the Imagine stream processor project and contributed to its VLSI implementation.

  • Diversity Redefined in the New Affirmative Action Era
    Examine the enormous challenge academic institutions face to ensure a diverse student population without giving preference to race or gender, according to the law. Dr. Juan E. Gilbert, professor and chair of the Human Centered Computing Division in the School of Computing at Clemson University, has attempted to address these issues with a data mining tool called Applications Quest. Applications Quest allows the use of race, gender or any other attributes to be considered in admissions, school assignments, employee hiring or any other application processing area, such that no preferences are given to race or gender.

  • DNA Self-assembly and Computer System Fabrication
    The migration of circuit fabrication technology from the microscale to the nanoscale has generated a great deal of interest in how the fundamental physical limitations of materials will change the way we engineer computer systems. The changing relationships between performance, defects, and cost have motivated research into so-called disruptive or exotic technologies. Chris Dwyer, from Duke University will present the theory, design, and methods of fabrication for DNA self-assembled nanostructures within the context of circuit fabrication. The advantages of this technology go beyond the simple scaling of device feature sizes (sub-20nm) to enable new modes of computation that are impractical under the constraints of conventional fabrication methods.

  • DNA Time Series Expression Data
    DNA microarray technologies enable researchers to measure the expression levels of thousands of genes simultaneously. More recently microarrays have been exploited to measure genome-wide protein-DNA binding events. Time series expression data offer particularly rich opportunities for understanding the dynamics of biological processes. Ziv Bar-Joseph presents algorithms for analyzing time series expression data at two different levels: individual genes and genetic regulatory networks.

  • Dynamic Invariant Detection
    This program begins with background on dynamic invariant detection, a technique for discovering invariants from execution traces, followed by a description of an application of program invariants to the task of refactoring, and then integrating two complementary techniques for manipulating program invariants.

  • Dynamics of Real-World Networks
    Emergence of the web and cyberspace gave rise to detailed traces of human social activity. This offers great opportunities to analyze and model behaviors of millions of people. Take, for example, an examination of planetary scale dynamics of a full Microsoft Instant Messenger network that contains 240 million people, with more than 255 billion exchanged messages per month (4.5TB of data), which makes it the largest social network analyzed to date. In this University of Washington program, guest speaker Jure Leskovec of Carnegie Mellon University, focuses on two aspects of the dynamics of large real-world networks -- dynamics of information diffusion and cascading behavior in networks, and dynamics of the structure of time evolving networks.

  • Efficient, Heterogeneous, Parallel Processing: The Design of a Micropolygon Rendering Pipeline
    Designing systems that are high-performance, power-efficient and easily programmable by non-experts is important at all levels of computing. Kayvon Fatahalian of Stanford University spoke at the University of Washington about this topic. While many recent innovations in parallel systems address this challenge in a general context, real-time graphics systems have achieved similar goals through domain-influenced co-design of algorithms, programming interfaces and heterogeneous, parallel hardware.

  • Electric Energy Systems
    Kai Strunz describes how real time digital simulators for energy systems are used for co-simulation that allows for an on-line exchange of information between real and virtual energy systems.

  • Electronic Election Results
    Electronic devices and systems potentially offer many of the same benefits to the process of conducting elections that they have already delivered to the worlds of finance and business. Unfortunately, the requirement for ballot secrecy, along with the high degree of complexity possible in today's devices, makes it impossible for the general voting population to directly infer that systems tasked with collecting and counting votes are behaving accurately and honestly. Andre Neff describes a new protocol, executable by voters while in the poll booth, that eliminates the need to trust vote casting software and hardware.

  • Embedded Links: A Misunderstood and Fundamental Element of Urban-Scale Networks
    Many urban communities have unequal access to Internet resources, presenting a technical challenge of providing a high-speed access infrastructure at an extremely low cost. To address this challenge, a first-of-its-kind, urban-scale wireless mesh network which provides Internet access to 1000's of users spanning multiple square kilometers in an underserved area in Houston, TX has been deployed. However, in this and other urban environments, IEEE 802.11 node interactions are affected by a vast array of factors including topology, channel conditions, modulation rate, packet sizes, and physical layer capture which are addressed in this lecture.

  • Embedded Networked Sensing Systems
    Embedded Networked Sensing Systems, commonly referred to as sensor networks, will form a critical infrastructure resource for society—they will monitor and collect information on such diverse subjects as tree canopies, endangered species, soil and air contaminants, medical patients and man-made structures. Deborah Estrin will describe several motivating science applications and the challenges in managing distributed data storage and processing within the Embedded Networked Sensing system.

  • Embedded Systems
    Embedded systems are ever present in our environment, and in future will be even more so. Resource management and optimization at all levels of abstraction in such embedded networks is of critical importance. Tajana Simunic Rosing addresses three different areas of embedded system resource management: management at the system and server level, hardware management and embedded software optimization. Simunic Rosing will also present a tool flow and a methodology that automates the use of complex processor instructions and pre-optimized software library routines using symbolic algebraic techniques.

  • Embedded Systems Capstone Design with Professor Gaetano Boriello
    A summary of student's final projects in this course. Projects include an automated home audio system called Migrating Audio, an in-building navigation system that allows you to call up a map of an unfamiliar building to help you find your way, and Virtual Neighbor, a system that remotely monitors a home or apartment to help you keep an eye on an elderly relative, for example.

  • Empowering the Individual
    Three technologies came together in 1994 to empower the creation of the modern Internet: the PC, high capacity disk storage for servers and a global high-speed network. Today powerful processors, vastly increased storage capacity and ever faster global networking are creating opportunities for a new generation of applications that will accelerate science, improve human communication, augment human memory and adapt everyday devices to the needs of the individual. In this colloquia, Rick Rashid, Senior VP at Microsoft Research examines the ongoing changes and will highlight technologies that demonstrate the potential for the future.

  • Energy Crisis, Smart Solutions
    Our nation’s electric grid must transform into an integrated digital system to meet expanding 21st century power demands. PNNL is a leading contributor in the nation’s billion-dollar push to develop “smart grids” and the technologies that will radically transform grid operation, increase energy efficiency, incorporate renewable energy, support electrification of the transportation sector and produce a smaller carbon footprint. Meanwhile, University of Washington engineers are inventing sensors to monitor resource use in real time and encourage efficiency in the home – smarter energy from source to user.

  • Energy Efficient System Design and Utilization
    Energy consumption of electronic devices has become a serious concern in recent years. Energy efficiency is necessary to lengthen the battery lifetime in portable systems, as well as to reduce the operational costs (e.g. cost of electricity) and the environmental impact (e.g. cooling fan noise) of stationary systems. Optimization in design and utilization of both hardware and software is needed in order to achieve more energy efficient systems.

  • Energy Minimization for Computer Vision via Graph Cuts
    Many energy minimization problems involve assigning a label to every node in a graph, where adjacent nodes should have similar labels. These optimization problems arise in fields as diverse as classifying web documents, biometry, maximum a posteriori estimation of Markov Random Fields, and computer vision. Until recently, these problems were thought to be completely intractable, since the energy functions have multiple local minima and the search space typically has thousands of dimensions.

  • Engaging Digital Natives in Information Technology Learning
    Innovative solutions are needed to ensure a new generation of digital natives is engaged in learning and becoming the future of IT management. In this program, Dick Nolan, professor emeritus of Harvard Business School, Robert Austin, professor at Copenhagen Business School and Shannon O'Donnell, PhD fellow at Copenhagen Business School, come together to discuss the future of digital natives.

  • Engineering the Tools of Scientific Discovery, Part 4 of 6
    Part four of the six part NAE Seattle Grand Challenges series dives deeper into eScience and technologies for the 21st century, with an introduction by Ed Lazowska, Director of the UW eScience Institute. Keynote speaker Larry Smarr, the founding director of the California Institute for Telecommunications and Information Technology at UCSD, addresses high performance cyberinfrastructure discovery tools for data intensive research. Following up with information on both data integration issues and social data collection are Catharine van Ingen, Partner Architect, eScience Group at Microsoft and Facebook data team member Johnathon Chang. Alon Halevy, Structured Data Management Research Head at Google, discusses removing the barriers to data analysis. The episode concludes with a lively Q and A session with all of the featured speakers.

  • Engineering Achievements of the 20th Century and Challenges for the 21st
    At the beginning of the 21st Century, it seems appropriate to reflect on what engineering has accomplished and, more importantly what challenges lie ahead. The impact of engineering on society, in the quality of the life of every citizen, has been immense over the last 100 years. The impact in the next century is likely to be even greater and to spread to a much larger fraction of the earth's population -- but that is not guaranteed. Significant challenges face the profession in ensuring that impact is positive, for example; these challenges raise issues that the profession has not faced before. Reducing unintended negative impacts of engineering is becoming increasingly critical. Similarly, spreading the benefits of engineering to more of the world's population is becoming critical. This lecture will reflect on these and other societal implications of engineering -- its past achievements and future challenges.

  • Engineering Better Medicines, Part 2 of 6
    In part two of the six part series chronicling the NAE Seattle Grand Challenges Summit, Matt O'Donnell, Dean of Engineering and John Markoff, correspondent for the New York Times, provide an introduction to the themes of "Engineering Better Medicines" and "Engineering Creativity." Keynote speaker Nicholas A. Peppas, of the University of Texas at Austin, discusses the involvement of engineers in developing solutions to medical challenges. He details the contribution of engineering in the development of new systems using genetic information, sensing small changes in the body, assessing new drugs and delivering vaccines at lower costs.

  • Engineering Better Medicines, Part 3 of 6
    Engineering better medicines for cancer treatment takes center stage in part three of the six part series chronicling the NAE Seattle Grand Challenges Summit. Keynote speaker Suzie Pun, Associate Professor in Bioengineering at the University of Washington, provides a history of chemotherapy and opportunities for improved treatment. She discusses how personalized medicines can lead to more specific delivery, improved drug efficacy and the ability to reach new disease targets.

  • Engineering Creativity: Opportunities in Global Health, Part 1 of 6
    Matt O'Donnell, Frank and Julie Jungers Dean of Engineering of the University of Washington provide an overview of the NAE Grand Challenges focused specifically on the topics of Engineering Better Medicines and Engineering the Tools of Scientific Discovery. Hugh Chang, director of Special Initiatives within the Office of the President, PATH, addressed the topic of "Engineering Creativity", specifically when it comes to the opportunities in global health.

  • Engineering the Tools of Scientific Discovery, Part 5 of 6
    In part five of the six part NAE Seattle Grand Challenges Summer series, Mike Griffin, Professor and Director of Center for System Studies, University of Alabama Huntsville, talks about unique examples of large system engineering projects, addressing both succeses and failures. Edward F. Crawley, Professor at Massachusetts Institute of Technology, shares why human space flight remains important and discusses the current goal of landing on Mars and potential permanent human presence. Bonnie Dunbar, President and CEO of the Museum of Flight contines the space exploration theme by addresssing the need to engineer tools that will contribute to continued universe exploration with both robotic and human explorers. The episode concludes with all speakers participating in a question and answer session.

  • Engineers Without Borders: Engineering with Soul
    Bernard Amadei, professor of Civil, Environmental and Architectural Engineering at the University of Colorado at Boulder, shares his passion to partner with disadvantaged communities to improve their quality of life through Engineers Without Borders. Engineers Without Borders is an international nonprofit organization that drives environmentally sustainable, equitable and economical engineering projects in disadvantaged communities worldwide while developing responsible engineers and engineering students. The central belief of the organization is "to build a better world, one community at a time."

  • Enhancing Creativity Through Toolkits
    Interface toolkits in ordinary application areas let average programmers rapidly develop software resembling other standard applications. In contrast, toolkits for novel and perhaps unfamiliar application areas enhance the creativity of these programmers. By removing low-level implementation burdens and supplying appropriate building blocks, toolkits give people a "language" to think about these new interfaces, which in turn allows them to concentrate on creative designs. This means that programmers can rapidly generate and test new ideas, replicate and refine ideas presented by others, and create demonstrations for others to try.

  • Ephemeral Instrumentation for Lightweight Program Profiling
    Program profiling is a useful mechanism for performance evaluation and code optimization. Profiling techniques that provide detailed information with extremely low overhead are especially important for systems that continuously monitor or dynamically optimize running executables. Described is an approach for program profiling called ephemeral instrumentation and show that it collects useful profiles with low overhead.

  • Error-Tolerant Networking Protocols
    Internet protocols were designed to tolerate failures, yet not all faults are alike. In this colloquia, David Wetherall reviews several protocols that proved vulnerable to software failures and describes improved designs that are less fragile.

  • Explore a Career in Paper Science Engineering
    How would you like to have most of your tuition paid for? And upon graduation know that you will have a job with a starting salary around $58,000 ? Then explore the PSE program at the University of Washington. This program applies natural products chemistry, chemical processing, and material science to the many uses of natural products and fiber based materials, including paper and biofuels manufacturing. The program also offers several grants to students and currently has 100% job placement.

  • Extended Static Checking
    Described is a system for detecting automatically at compile time certain errors that are normally not detected until run time, and sometimes not even then. For example, array bounds errors, NIL dereferences, and race conditions and deadlocks in multi-threaded programs. The system has been implemented both for Modula-3 and for Java.

  • Eye on the Universe: Final Mission to Hubble
    The 19-year-old Hubble Space Telescope has yielded stunning images and a remarkable scientific legacy – revealing new insight into the age of the universe, black holes and the role of "dark energy" in our expanding universe. University of Washington alum Gregory Johnson piloted the space shuttle Atlantis for the final service mission to Hubble. Imagine the extreme challenges of launching the shuttle 358 miles into space, capturing the huge telescope and making tricky repairs during five spacewalks. Johnson takes us on a thrilling journey into space and inside the final mission to Hubble.

  • Five Forces in the Network Economy
    Information goods -- from movies and music to software code and stock quotes -- have supplanted industrial goods as the key drivers of world markets. Confronted by this New Economy, many instinctively react by searching for a corresponding New Economics to guide their business decisions. Executives charged with rolling out cutting-edge software products or on-line versions of their magazines are tempted to abandon the classic lessons of economics, and rely instead on an ever-changing roster of trends, buzzwords, and analogies that promise to guide strategy in the information age.

  • Fleet, Infinity and Marina
    Simplicity isn’t a word usually associated with computer chips. For Ivan Sutherland though, it may provide the key to streamlining chip design and advancing program communication. In this video from the University of Washington, Sutherland, vice president of Sun Microsystems and founder of the Asynchronous Research Center at Portland State University, discloses his three, radically pared-down computing architectures named Infinity, Marina and Fleet.

  • Fluid Interaction for High Resolution Wall-Size Displays
    Guimbretiere explores how to bridge the gap between the power provided by current desktop computer interfaces and the fluid use of whiteboards and pin-boards. Observing fluid expert interactions in everyday life, such as driving a car or playing a violin, his team has designed, and built a fluid interaction framework which encourages gesture memory, reduces the need for dialog with the user, and provides a scoping mechanism for modes. To validate their design, they built the Stanford Interactive Mural, a 9 Mpixel whiteboard-size screen, evaluated the performance of their proposed menu system FlowMenu, and implemented PostBrainstorm a digital brainstorming tool. PostBrainstorm lets users gather and organize sketches, snapshots of physical documents, and a variety of digital documents on the Interactive Mural. PostBrainstorm was very well received by professional designers during testing, and was used in the early design stage of the Chrysler Design Award 2001. It demonstrates the feasibility of fluid, transparent interactions for complex, real life applications.

  • From Badware to Malware: Taming the Malicious Web
    Today's Web is where everything happens: we work, we play, we live our lives. Unfortunately, criminals also use the World Wide Web as a resource for stealing and abusing the very information made accessible by the Web via malware installations.

    In his lecture at the University of Washington, Giovanni Vigna addresses one of the latest trends in Web-based malware – the leveraging of legitimate Web sites for the delivery of attacks that target vulnerabilities in client-side software. How can we make Web applications more secure? How can we detect and block attacks against the components of the Web?



  • From GUIs to PUIs
    Computers will not always be used as glorified typewriters - the future of human-computer interaction is not in typing words and dragging icons. How are we to interact with technology as it becomes more powerful and ubiquitous? How can we make technology truly user-friendly and accessible to all? We need to move toward "perceptual user interfaces" (PUIs) - highly interactive, multimodal interfaces that are modeled after natural human-to-human interaction. This requires better understanding of human perceptual capabilities as well as improved machine perception. Most importantly, it requires researchers from various disciplines, including vision, speech, UI, learning, haptics, cognitive psychology, and others, to cooperate deeply.

  • From Perception and Discriminative Learning to Interactive Behavior
    A strong symbiosis lies between machine learning and machine perception. Just as we learn to reason and interact with the world through our senses, a smart sensing system could acquire data to drive higher level learning problems. Ironically, learning and probabilistic methods themselves can provide the driving machinery for perception as well. I demonstrate several examples of probabilistic sensors in wearable and room-based environments. These human-centered systems perform object detection, face tracking, 3d modeling, recognition, and topic-spotting in real-time.

  • Functional Image Synthesis
    Pan is a declarative language and optimizing compiler for image synthesis, based on a simple but precise semantic model: pictures are functions from infinite, continuous space to colors with partial opacity; and effects are functions over pictures. Because of the centrality of functions, Pan is based on the functional programming paradigm, and is in fact embedded in the functional language Haskell.

  • Garbage Collection Algorithms
    The increasing reliance on garbage collected languages such as Java and C++ requires high performance from the memory management system. Kathryn McKinley presents two new garbage collection algorithms, Beltway and Ulterior Reference Counting, to improve performance.

  • Gender, Lies and Video Games: Women and Computer Sciences
    Princeton University professor, Maria Klawe, explores how girls and women differ from boys and men in their uses of, and attitudes towards, computers and computing. From playing computer games to pursuing computing careers, the participation of females in the computer sciences tends to be very low compared to that of males. Klawe discusses research findings on this issue and initiatives designed to increase the participation of females in computing.

  • Generalized Alias Analysis on a Basis for Programming Tools
    Tools for automatic program analysis promise to improve programmer productivity by searching and summarizing large bodies of code. However, the phenomenon of aliasing -- different names being used to refer to the same data -- places fundamental limits on the effectiveness of simple textual analyses. My thesis system, Ajax, addresses this problem by incorporating static alias analysis into a number of different tools to aid Java programmers. The system handles arbitrary Java programs and has been evaluated against real Java applications such as the Javac Java compiler and the Javafig graphic editor.

  • Genome: Transcriptional Regulatory Modules
    Press coverage of the release of the human genome sequence in 2001 focused on the surprisingly small number of genes in the human genome (approximately 30,000), possibly explained by the complexity of the switching machinery involved in activating and deactivating genes in various environmental conditions. In this program, William Noble describes a technique for locating occurrences of the transcriptional regulatory modules that act as switches.

  • Gone Phishing
    If you use e-mail, chances are you’ve been touched by phishing. This type of Internet fraud uses spam to gain access to private information. Messages might contain information about unauthorized activity on your credit card and direct you to update your records. This student documentary reveals the scams Internet phishers commonly use to steal account information — or even steal your identity.

  • Google Ad Systems
    Billions of dollars (and euros and yen, too) flow through Google Ad Systems. In this talk, Narayanan Shivakumar, a Google distinguished entrepreneur and director of Google's Seattle-Kirkland R and D center, gives us a peek under the covers and examines how a combination of auction theory, machine learning and large systems can power successful businesses like Adwords and Adsense.

  • Google: A Behind-the-Scenes Look
    Search is one of the most important applications used on the Internet and poses interesting challenges in computer science. Providing high-quality search requires understanding across a wide range of computer science disciplines. In this program, Google Fellow Jeff Dean describes some of these challenges, discusses applications Google has developed, and highlights systems they've built, including GFS, a large-scale distributed file system, and MapReduce, a library for automatic parallelization and distribution of large-scale computation. He also shares observations derived from Google's Web data.

  • Grounding the Lexical Semantics of Verbs in Visual Perception using Force Dynamics and Event Logic
    An implemented system, called Leonard, that classifies simple spatial motion events, such as 'pick up' and 'put down', from video input is presented. Unlike previous systems that classify events based on their motion profile, Leonard uses changes in the state of force-dynamic relations, such as support, contact, and attachment, to distinguish between event types.

  • Group 01: Computer Tech and Learning Disabilities
    "People with Disabilities and Computer Technology" addresses adaptive technology and computer applications for people with disabilities. In "Computers and People with Learning Disabilities," people with disabilities demonstrate the wide array of current technology they use for school and work.

  • Guidelines: Web Data Collection for Understanding and Interacting with Your Users
    The global growth of the World Wide Web challenges technical communicators to reconsider the methods we use to create designs that meet the goals and needs of our users. This presentation focuses on taking advantage of the Web's potential for interactivity between designers and users. It offers strategies for getting data from users of Web sites and using it for two main purposes: (1) analyzing audience and patterns of use to support continuous redesign, and (2) building a relationship or sense of community on a Web site.

  • Hancock: A Language for Computing with Large Data Streams
    Like many companies, AT&T collects and analyzes data to help run its business. What makes our situation unusual is the scale (and variety) of data collected. For data analysts, who are highly-trained statisticians but not necessarily experienced programmers, the scale of the data can present a significant barrier.

    The Hancock project started three years ago with the hypothesis that the key to reducing the complexity of computing with massive data sets was to identify a set of essential abstractions and incorporate them into a language.

  • Hardness Amplification by Repetition
    Does computing k times as many functions require k times the computational effort? In this talk, we discuss a few scenarios in which variants of this question have been studied. This talk will examine hardness of approximation, communication complexity and spherical cubes.

  • Hardware Security Mechanisms for Authentication and Trust
    Explore novel lightweight hardware-based mechanisms for ensuring security, intellectual property (IP) protection and trust of integrated circuits (ICs) and systems with Farinaz Koushanfar of Rice University.

    New security methods are in demand due to the proliferation of the fabless semiconductor business model, increase of third-party IP reuse, emergence of personal security devices and the high overhead of traditional cryptographic protocols for embedded systems.

    Active hardware metering is a first system of security mechanisms and protocols that enable the design house to gain active post-fabrication control of each produced IC, their properties and terms of use or by run-time disabling of ICs in case of tamper detection.

    Koushanfar also shares his ongoing work in security analysis, safeguarding, implementation and the fabrication of new families of physical unclonable functions, and their use in secure system design. He also discusses attacks and countermeasures.



  • HDTV over Internet: Bag Lunch Panel
    The UW and Internet HDTV Computing & Communications involvement in developing and integrating technology and techniques for the distribution of high-quality video over the Internet.

  • Herald: Global Event Notification
    Mike Jones of Microsoft Research discusses the design philosophy and initial design decisions of Herald, a highly scalable global event notification system that is being designed and built at Microsoft Research.

  • High-Dimensional Computational Geometry
    Computing with massive and high-dimensional data is critical to a large and diverse set of applications, including multimedia and hyperlinked databases (the World Wide Web being the prime example), data mining, machine learning, computational statistics, and vector quantization/compression. Improving performance in the above applications has been an important research goal in a variety of fields, including Computational Geometry and Databases. Unfortunately, the running times of the algorithms discovered so far depend exponentially on the dimension, which usually makes them inefficient in the aforementioned applications.

    It is possible to overcome this "curse of dimensionality" and obtain algorithms that are efficient in theory and practice, as long as one is satisfied with approximate answers.

  • High-Tech Entrepreneurship: An Idiosyncratic View of Technology
    Entrepreneur and technology investment professional W. Hunter Simpson offers his view of the future of technology in the Puget Sound area. The High-Tech Entrepreneurship Series is sponsored by the University of Washington Program in Entrepreneurship and Innovation, Department of Computer Science and Engineering, American Electronics Association, Washington Software Alliance, and the Washington Biotechnology and Biomedical Association. The series is made possible, in part, by grants and funding from the Herbert B. Jones Foundation and Wells Fargo.

  • How People Treat Interfaces Like People: Social Psychology and Design
    A series of recent experimental studies that demonstrate the breadth and depth of social responses to computers and interfaces are discussed. Specifically, Cliff demonstrates that one can directly apply theories and methods from social psychology (concepts will include similarity-attraction, consistency, social identity, and attribution theory) to human-computer and human-web interaction.

  • How to Get the Most Out of a Startup
    Glenn Kelman, Redfin CEO and Plumtree Software founder, talks to computer science undergraduates about how to evaluate, interview at and negotiate with startups and also how to think about some day starting a company of their own. Kelman gives tips for dealing with startups going into recruiting season.

  • Identity-Based Encryption: A Biometric Approach to Privacy
    Identity-Based Encryption (IBE) allows the use of an identity as a public key. For example, to encrypt a message one would only need to know the recipient's identity, the string 'bob@yahoo.com', without needing access to a traditional public key infrastructure. In this program, Brent Waters presents a new form of IBE that uses human biometrics as identities (public keys), discussing the motivations for biometric-based IBE, the primary challenges in developing such as system, and a cryptographic solution called, Fuzzy Identity-Based Encryption.

  • Image Segmentation using Spectral Rounding
    This lecture presents a new image segmentation algorithm, Spectral Rounding (SR), and a fast solver used for segmenting 2D images. The second issue addressed is fast algorithms for finding the associated eigenvectors and solving related linear systems.

  • Improving Information Interactions
    Whether reading email, searching the web or mining data, we are constantly interacting with information. Melody Ivory highlights research efforts aimed at improving a subset of these information interactions: automated tools for improving web site designs, approaches for improving search tasks and potentially reducing email overload.

  • In Search of the Joule of Computer Architecture
    This program outlines some preliminary work in the search for a fundamental unit of measurement for computer architectures.

  • Inductive Learning and Representation for Text Categorization
    As the volume of electronic information increases, there is growing interest in developing tools to help people better find, filter, and manage these resources. Text categorization - the assignment of natural language texts to one or more predefined categories based on their content - is an important component in many information organization and management tasks. Today, most text categorization is done by people. Human categorization is very time-consuming and costly, thus limiting its applicability especially for large or rapidly changing collections. Consequently there is growing interest in developing technologies for (semi-)automatic text categorization.

  • Information Fusion: Multidocument Summarization
    The number and variety of online news sources makes it difficult for people to track the news concerning even a single event. The focus of this talk by Regina Barzilay of Columbia University is information fusion, a technique which identifies redundant information and synthesizes a coherent summary.

  • Information Technology Leadership Learning in Action
    Dick Nolan, professor emeritus of Harvard Business School, and Robert Austin, professor at the Copenhagen Business School, discuss how to train the next generation in IT management: Train today’s leaders to engage digital natives, for whom technology is an essential part of the world. With that being said, is active learning the right approach?

  • Integrated Land Use, Transportation, and Environmental Modeling: Recent Research and Future Directions
    The purpose of this program is to provide a tool for citizens' groups, urban planners, elected officials, and others to help predict future patterns of urban development and impact under different possible scenarios over periods of 20 or more years. The talk will include an overview of the modeling activity, followed by a technical discussion of the system and of particular computational aspects, including designing a software architecture to support relatively independent implementation and evolution of the different component models, automatically choosing appropriate visualizations, and supporting new model development with a domain-specific programming language. It concludes with a discussion of current and future research directions.

  • Intelligence in Wikipedia
    Berners-Lee's vision of the Semantic Web is hindered by a chicken-and-egg problem, which can be best solved by a bootstrapping method: creating enough structured data to motivate the development of applications. UW CSE believes that autonomously Semantifying Wikipedia is the best way to bootstrap. They choose Wikipedia as an initial data source, because it is comprehensive, high-quality, not too large, and contains enough manually-derived structure to bootstrap an autonomous, self-supervised process. This talk will present their success to date in this endeavor.

  • Intelligent Tutoring Systems
    Intelligent Tutoring Systems (ITS) are computer-based instructional tools that rely on artificial intelligence techniques to generate individualized interactions tailored to a student's learning needs. Cristina Conati discusses how the scope and effectiveness of ITS can be increased by extending the range of features captured in a student model to include domain independent, meta-cognitive skills and affective states.

  • Interaction Techniques for Ambiguity Resolution in Recognition-based Interfaces
    Because of its promise of natural interaction, recognition is coming into its own as a mainstream technology for use with computers. Both commercial and research applications are beginning to use it extensively. However the errors made by recognizers can be quite costly, and this is increasingly becoming a focus for researchers. We developed OOPS, a toolkit that supports resolution of input ambiguity through mediation.

  • Interactive Visual Media
    Anandan outlines ongoing research at MSR that is aimed at enabling interaction with visual media and people to people interactions via visual media. The technologies support image and video based 3D scene modeling, video processing and enhancement, personal photo and video sharing, annotation, and organization, video understanding, and novel adapative approaches for document layout and for presentations.

  • Internet Congestion Control, Bandwidth-Delay Product

    Katabi explains that theory and experiments show that as the per-flow product of bandwidth and latency increases, TCP becomes inefficient and prone to instability, regardless of the queuing scheme. This failing becomes increasingly important as the Internet evolves to incorporate very high-bandwidth optical links and more large-delay satellite links.

    To address this problem, a novel approach to Internet congestion control that outperforms TCP in conventional environments, and remains efficient, fair, scalable, and stable as the bandwidth-delay product increases was developed. This new eXplicit Control Protocol, XCP, generalizes the Explicit Congestion Notification proposal (ECN). In addition, XCP introduces the new concept of decoupling utilization control from fairness control. This allows a more flexible and analytically tractable protocol design and opens new avenues for service differentiation.

  • Internet Evolution and Some Challenges for the Early 21st Century
    Although the Internet has been around for 35 years in concept and 25 years since roll out, there are still many capabilities that could improve its utility. Broadcast models of operation; dealing with mobility and multi-homing, coping with persistent connections, accommodating strong end/end authentication, expanding the address space, dealing with multilingual domain names, implementing DNSSEC, supporting an interplanetary extension of the Internet, adding more security to all layers and a host of other features. These topics will be explored in this talk.

  • Internet Search Engines
    Computer Science & Engineering Distinguished Lecturer, Udi Manber, Amazon.com, talks about recent search engine innovations and the future of searches on the Internet.

  • Internet Telephony: Will it Kill the Telephone Companies, the Internet, or Both?
    Internet telephony differs from other Internet applications in an important way: there is a pre-existing industry out there that thought it was already selling that service. There is no "e-mail" company or "World Wide Web" company that had pulled a wire to your house to sell you service. The collision between the Internet and the phone companies will be a definitive event in shaping the Internet over the next decade. Neither party will survive intact.

    The first step in sorting out this collision is to distinguish among the wide range of services that have been called "Internet telephony".

  • Introduction of the Mobile Robotics Lab
    We introduce the new denizens of the mobile robotics lab and give an overview of our software system for mobile robot control.

    Some of the key problems in mobile robotics arise from the inherent uncertainties involved in sensing and acting in the real world. Our control system makes extensive use of probabilistic methods in order to deal with these uncertainties. We focus on probabilistic methods for mobile robot localization and map building. The first problem is the estimation of a robot's position within its environment, and the second problem is the task of building a model of a robot's environment using sensor information collected by the robot. We introduce and demonstrate the application of efficient Monte Carlo methods to these problems. Furthermore, we discuss directions for future research.



  • Introduction to Venture Capital
    Securing financing is a critical step for most startup companies as they progress from an initial idea to a fully-staffed company shipping real products. For an entrepreneur starting a company, having a clear understanding of potential financing mechanisms can be just as important as knowledge of the technology and market opportunities. Examine startup financing in general and venture capital in particular in this video from the University of Washington’s Computer Science and Engineering Department.

  • Introduction to VLSI Design with Professor Chris Diorio
    A summary of student's final projects in this course. Students in this class focused on chip design. One of the projects described is the design of an elevator controller.

  • It's Not Just About Tuskegee: The History of African Americans and Medicine
    The Tuskegee Syphilis Study is on of the most well known events in the history of African Americans and medicine. Dr. Vanessa Northington Gamble provides an overview of how issues of race and racism continue to influence the development of American medicine and public health. Gamble outlines a historical framework for understanding the contributions and experiences of African-American healthcare providers and the experiences of African-American patients.

  • Keynote by William Gates Sr. and Panel: “Around the World with EWB”

    Keynote address by William Gates, Sr.: "Sustaining Engineering and Global Health"

    Panel discussion: "Around the World with EWB"

    Engineers Without Borders is an international nonprofit organization that partners with disadvantaged communities worldwide to improve their quality of life through environmentally sustainable, equitable and economical engineering projects.

    In the conference's keynote address, William Gates Sr., Co-Chair of the Bill & William Gates Foundation, presents “Sustaining Engineering and Global Health.”He discusses efforts that the Gates Foundation is currently making to help fund engineering and health projects throughout the world. Following the keynote address, learn how Puget Sound students and professional engineers are implementing development projects “Around the World.”. The University of Washington’s Donee Alexander, a Ph.D. student in Civil and Environmental Engineering, discusses a recent project by the UW student chapter to improve water supply and cook stoves in Yanayo, Bolivia. The Seattle University student chapter of EWB and Puget Sound professional chapter also present their groups' recent initiatives in Thailand and Ethiopia.

  • Learning Hierarchical, Nonparametric Models for Visual Scenes
    Computer vision systems use image features to detect and categorize objects in visual scenes. In this University of Washington program, learn about Erik Sudderth MIT/UC Berkeley research that explores hierarchical models using contextual and geometric relationships for more effective learning from large, partially labeled image databases.

  • Learning Nonlinear Data Manifolds
    How many numbers does it take to represent a complex object such as an image of a face? Obviously, one number per pixel is enough, but many fewer are actually needed. In fact, there is a thin "submanifold" of faces hiding in the very high dimensional image space. Learning the structure of such manifolds is the problem of nonlinear dimensionality reduction. Its solution allows compression, generation, interpolation, and classification of complex objects.

  • Learning to Map Sentences to Meaning
    Building automated systems that participate effectively in natural language conversations is one of the classic goals of research in artificial intelligence. Machine learning methods hold significant potential for addressing many of the challenges involved with these systems. This talk describes machine learning algorithms for the problem of mapping natural language sentences to representations of their underlying meaning.

  • Learning to Program DNA
    Ongoing improvements in DNA sequencing and DNA synthesis technologies are making genetic material (physical DNA molecules) and genetic information (DNA sequence data) interconvertible. For example, researchers have demonstrated the construction of DNA molecules up to 7,700,000 base pairs, a length that is long enough to encode all known viruses, most important bacteria, and almost the entire genome of S. cerevisiae (baker's yeast).

  • Learning, Logic, and Probability
    Artificial intelligence systems must be able to learn, reason logically, and handle uncertainty. Research has focused on each of these goals individually, and only recently have attempts been made to achieve all three at once. In this colloquia, Pedro Domingos, UW Computer Science & Engineering, describes Markov logic: a representation that combines the full power of first-order logic and probabilistic graphical models, and algorithms for learning and inference in it. Experiments in a real-world university domain illustrate the promise of this approach.

  • Lessons Learned from Applying Control Theory to Computing Sytems: A Manifesto for Resource Management Engineering
    Joe Hellerstein summarizes his experience over the last 5 years with applying control theory to resource management solutions in computing systems. His experience at IBM and Microsoft has been that problems of dynamics can often be addressed by using a simple set of techniques based on discrete time, linear, time-invariant.

  • Life-Sized Learning
    In this distinguished lecturer colloquia, Leslie Pack-Kaelbling describes the combination of techniques, from machine learning to operations research, that have allowed major advances in learning and planning for uncertain environments.

  • Linear Time Encodable/Decodable Codes
    Venkatesan Guruswami presents a simple, explicit construction of error-correcting codes that have a near-optimal rate vs. error-correction trade-off and are encodable and decodable in linear time. These codes nearly match the "Singleton bound" and thus their rate vs. distance trade-off is essentially optimal, and so is their encoding/decoding time (up to constant factors). Previously known codes over a constant-sized alphabet that approached the Singleton bound (eg. certain algebraic-geometric codes) suffered from complicated constructions and/or high (certainly, super-linear) encoding/decoding complexity.

  • List Decoding
    Venkat Guruswami's research focuses on error-correcting codes, graph-theoretic optimization and approximation problems, probabilistically checkable proofs, hardness of approximations and complexity theory. This colloquia is a survey into the subject of list decoding in which Guruswami provides an overview of the substantial progress on code constructions and the algorithmic front.

  • List Decoding of Error-Correcting Codes
    Error-correcting codes are combinatorial objects designed to achieve reliable transmission of information on a noisy channel. A fundamental algorithmic challenge in coding theory and practice is to efficiently decode the original transmitted message even when a few symbols of the received word are in error. The naive search algorithm runs in exponential time, and several classical polynomial time decoding algorithms are known for specific code families. Traditionally, however, these algorithms have been constrained to output a unique codeword.

  • Location Enhanced Web Services
    Researchers from Intel, UW, UCSD, and elsewhere have recently created an initiative to encourage the broad adoption of location-enhanced web services. The consensus is that to be widely adopted, location-aware computing must be as effortless, familiar, and rewarding as web search tools like Google. Therefore, the research collaboration is producing open software that enables private, course grain, indoor and outdoor positioning on cellular mobile computers with no additional hardware.

  • Location Estimation for Activity Recognition
    Dieter Fox illustrates how Bayesian filtering can be applied to estimate the location of a person using sensors such as GPS, infrared, or WiFi. The techniques track a person on graph structures that represent a street map or a skeleton of the free space in a building. In the context of GPS, the research indicates how a graph representation can be embedded into a hierarchical activity model that learns and infers a user's daily movements through the community.

  • Logic in Computer Science
    During the past thirty years there has been extensive interaction between logic and computer science. The argument is that logic plays a fundamental role in computer science, similar to that played by calculus in the physical sciences and traditional engineering disciplines. Moshe VardiÕs non-technical discussion will provide an overview of the unusual effectiveness of logic in computer science by surveying its history, dating back to the time of Aristotle and Euclid, and showing how logic actually gave rise to computer science.

  • Logical Inference Systems
    The ability to make logical inferences automatically is a central problem in a number of areas of computer science, ranging from AI to proof complexity to formal methods in software engineering and computer-aided design. Paul Beame focuses on the current state of propositional inference and the challenges and opportunities that lie ahead.

  • Mac OS X for UNIX Users
    The underlying operating system in Mac OS X (Jaguar) actually has a long history stretching back to BSD UNIX and CMU Mach. Bud Tribble will cover its interesting evolution and its place in the current line-up of mass market operating systems. Topics include: the miracle (and challenges) of actually providing ease of use on top of BSD, the interaction of Mac OS X with the open-source software community (Darwin) and recent developments in Mac OS X (Rendezvous networking, Safari browser, X Window Server support, OpenGL, Java tools).

  • Machines Reasoning About Machines
    Computer hardware and software can be described precisely in mathematical logic. Mechanical theorem proving techniques can be used to reason about those descriptions. But how practical is this vision? This program describes how one particular theorem prover, ACL2, is being applied to industrial verification projects. ACL2, developed by Matt Kaufmann and J. Strother Moore, is a descendent of the Boyer-Moore theorem prover.

  • Machines with Emotional Intelligence
    Over 70 studies on human-machine interaction in the last decade have pointed to an intriguing phenomenon: People interact with machines in a way that is basically social, even when the interaction was not designed to be that way. This program will describe how we're giving computers some social skills, specifically the ability to recognize and respond appropriately to human emotion. Examples are shown on keyboard-mouse-monitor systems that try to assess user frustration for usability feedback, and wearable systems that classify affective state based on skin-surface measurements.

  • Making Faces: A Technique for Realistic Facial Animation Capture
    Realistic facial animation could prove useful for a variety of applications, such as teleconferencing and customized user-interface agents. Unfortunately, easily produced, believable animation has proven to be an elusive goal. Humans are experts in reading faces and in noticing even small, subtle errors in an animation. The appearance of the face results from a complex interaction of muscles, bone, tissue and skin, making it difficult to simulate to the necessary level of detail. To circumvent this problem, we propose capturing live, high fidelity animation, storing it in a database and using the database to generate new combinations of speech and expressions.

  • Market-Making: From Algorithms for Price-Setting to Emergent Market Properties
    With the dramatic increase in electronic exchanges and automated trading in recent years, it has become important to develop new computational and algorithmic tools for analyzing market properties and designing software agents that participate in market activities. This program presents a Bayesian algorithm that can be used by a market-making agent to continuously post prices and update its beliefs based on the sequence of trades it sees. This algorithm leads to an interesting characterization of the market- maker's exploration-exploitation dilemma as a tradeoff between price discovery and profit-taking. It also allows for the building of richer agent-based models of markets that can be useful both in understanding properties of existing markets and in predicting the impacts of structural changes.

  • Massive Parallelism in the TeraOPS Chip
    Ambric, a Portland semiconductor startup, believes the key to massive parallelism in a high performance programmable chip is to define the right programming model first, then develop silicon architecture and tools to implement that model. In this talk, Mike Butts, Ambric, Inc., talks about Ambric's new chip that harnesses hundreds of 32-bit CPUs and memories in an power-efficient asynchronous system that is sensible to program and delivers up to one teraOPS performance.

  • Mathematical Sculpture in Stone and Bronze
    This is the keynote address for the Millennial Open Symposium on the Arts and Interdisciplinary Computing, sponsored by the University of Washington Animation Research Labs.

    Helaman Ferguson is both a mathematician and a sculptor: one of his algorithms was listed in the top ten algorithms of the century and his sculpture can be found in institutions of higher learning and research worldwide. In the golden age of mathematics that we live in Helaman celebrates theorems as among our civilizations crown jewels. With cutting edge technology he carves billion year old stone with computer virtual image projection to create warm and human forms embodying timeless mathematical ideas. Claire Ferguson will discuss her book: Helaman Ferguson: Mathematics in Stone and Bronze.



  • Measuring the Accuracy of Distributed Algorithms on Multi-Robot Systems
    Distributed algorithms running on multi-robot systems rely on ad-hoc networks to relay messages throughout the group. The propagation speed of these messages is large, but not infinite, and problems in algorithm execution can arise when the robot speed is a large fraction of the message propagation speed. In this work, we focus on measuring the accuracy of multi-robot distributed algorithms.

  • Mechanism Design: Private Value Optimization
    The design of protocols for resource allocation and electronic commerce among parties with diverse and selfish interests has spawned a great deal of recent research at the boundary between economics, game theory and theoretical computer science. Recent work in this area has drawn on ideas from the field of mechanism design, or 'incentive engineering'. In this colloquia, Anna Karlin surveys recent work in this area, with a specific focus on generalized auction design and analysis, and presents some directions for future research.

  • Meshes and Geometry Processing
    Progress in shape acquisition technology (laser scanners, etc.) has resulted in an abundance of large meshes representing highly detailed geometry. The acquired, densely sampled models are used in many application areas ranging from engineering design to entertainment and computational science.

    Processing of geometric information stored in such large meshes is the main theme of this talk.

  • Metric Clustering
    One measure of the quality of a k-clustering is the sum of all pair-wise distances inside the clusters, which must be minimized. Lecturer Claire Kenyon discusses techniques and algorithms, first for the complementary problem, which can be seen as a metric analog of Max-Cut in graphs, then for 2-clustering, and finally sketch extensions to variants with other objective functions or with cardinality constraints. The algorithms are based on random sampling.

  • Metric Geometry and Computer Science
    The combinatorial landscape is fraught with complexity. Computing optimal solutions to natural problems is often intractable, and understanding discrete systems can be difficult without the presence of richer structures to guide our intuition. In this talk, James Lee focuses on two examples. The first involves low-distortion embeddings of finite metric spaces, and its application to algorithms and structural theorems for cuts, flows, and balanced separators in graphs. The second revolves around abstract notions of dimension and how they can be used to characterize the complexity of certain problems on metric spaces and networks.

  • Microsoft's Parallel Computing Platform: Applied Research in a Product Setting
    The goal of Microsoft's Parallel Computing Platform (PCP) team is to enable the shift to modern, multi- and manycore hardware, by providing a runtime, programming models, libraries, and tools that make it easy for developers to construct correct, efficient, maintainable, and scalable programs through the use of parallelism. In doing so, tens of years of industry research has been combined and applied in a myriad of ways. This talk examines PCP's current progress, explicitly relating it to specific research of the past and present, in addition to surveying future efforts and possible research opportunities.

  • Mira and the Wind
    UW Student teams created this animated short in a Computer Animation Capstone Design course taught by Barbara Mones. The short demonstrates how a young girl comes to terms with death of her grandfather.

  • Model Checking Software Artifacts
    Model checking techniques have proven to be an effective means for finding subtle defects in hardware designs and communication protocols. In this talk, Matthew Dwyer describes Bandera, a toolset for model checking Java programs, and how our experiences developing and using the system have evolved. The Cadena toolset will also be described: it supports the definition of event-driven component-based designs, model checking properties of those designs, and checking the conformance of component designs and implementations.

  • Model Management: Databases


  • Modeling, Analysis and Optimization of On-chip Communication Architectures
    Traditionally, design space exploration for Systems-on-Chip (SoCs) has focused on the computational aspects of the problem at hand. However, as the number of processing elements on a single chip and their performance continue to increase, the design of the communication architecture plays a central role in defining the area, performance and energy consumption of the overall system. Furthermore, the global interconnects cause unpredictable delays and high power consumption. To mitigate these kinds of effects, the network-on-chip (NoC) communication architectures have emerged recently as a promising alternative to the classical bus-based and point-to-point communication architectures.

  • Modern Asynchronous Circuit and System Design
    Asynchronous circuits and systems do not use a global clock to coordinate their activities. Because they are not synchronized to a global clock they can offer significant advantages over traditional clocked systems in a variety of applications. Recent practical advances in asynchronous circuit and system design have resulted in renewed interest by designers. As a result, asynchronous systems are being viewed as an increasingly viable alternative to globally synchronous system organization.

  • Modern Views: A Conversation on Northwest Modern Architecture: Preview
    Five University of Washington alumni offer their perspectives on Northwest Modernist architecture. Architects Arne Bystrom, Gene Zema, Fred Bassetti, Wendell Lovett and Ralph Anderson shaped this regional style of architecture in the mid-20th century, adapting modernism to suit the unique needs of the Pacific Northwest. This preview offers a look at the documentary by the University of Washington Department of Architecture and studio/216.

  • Multi-robot Exploration
    Efficient exploration of unknown environments is a fundamental problem in multi-robot coordination. As autonomous exploration and map building becomes increasingly robust on single robots, the next challenge is to extend these techniques to large teams of robots. Dieter Fox provides an overview of research into multi-robot exploration and mapping, developed within the CentiBOTS project.

  • Multiprocessor Architectures for Programmability
    Luis Ceze believes the main problem confronting computer architects today is designing computer systems that help simplify parallel programming. Ceze presents two powerful computer architecture primitives that help with this. Together, these two techniques offer promising directions in the critical area of novel multiprocessor architectures for programmability.

  • Natural Language Processing
    Natural language processing offers a rich problem domain for machine learning approaches. Many NLP problems require the induction of a mapping that involves complex, discrete structures such as strings, labeled sequences, or trees. In this distinguished lecture, Michael Collins describes how 'large margin' methods in machine learning can be generalized to 'structured' problems found in NLP.

  • Netpoints: Volume 1
    Executives in the Internet industry present several topics: Sean Paul of Aventail discusses virtual private networks; Jeff Schrock, CEO of Activate, talks about webcasting; Scott Whitcomb, director of broadcast engineering at Activate explores streaming media; Cindy Ball, Founder and CEO of Toonedin, presents on shockwave/flash; and Dr. David Anderson, Project Director of SETI@Home, explains distributed computing.

  • Netpoints: Volume 4
    The five NetPoints in Volume 4 include the following: "Accessibility for the Disabled," by Dan Comden, Adaptive Technology Consultant; "UW Adaptive Technology Lab; Project Management and Team Relations," by Julie Hillers, director Of Project Management; "Working on a Team," by Angelina Churches, Freelance Designer, AngelinaLTD.com; "Internet 2," by Steve Corbato, director, Backbone Network Infrastructure, University Corporation for Advanced Internet Development (UCAID); "Genetics and Computing," by Professor Maynard Olson, Director UW Genome Center.

  • Network Coded Wireless Architecture
    Wireless is becoming the preferred mode of network access. The performance of wireless networks in practice, however, is hampered due to the harsh characteristics of the wireless medium: its shared broadcast nature, interference, and high error rate. Traditionally, network designers have viewed these characteristics as problematic, and tried to work around them. This talk will show how we can turn these challenges into opportunities that we exploit to significantly improve performance.

  • Network Performance Routing Strategies
    In large networks, it can be difficult or even impossible to impose optimal routing strategies on network traffic. On the other hand, permitting network users to act according to their own competing interests precludes any type of global optimality, and therefore carries the cost of decreased network performance. Roughgarden discusses methods for quantifying the worst-possible loss in network performance arising from such non-cooperative behavior, also briefly covering methods for improving the non-cooperative solution, including network design and edge pricing.

  • Networking at Home - Directions in Connected Computing for the Consumer
    A variety of forces are converging that will significantly impact computing as seen by consumers. First, the advent of inexpensive PCs coupled with the natural development of a replacement market is increasing the number of homes with more than one PC. Second, broadband Internet connections are beginning to appear in volume for residential use. Third, networking technologies optimized for use by the consumer are emerging. Finally, computing capabilities and the Internet are penetrating many existing consumer devices as well as creating opportunities for new ones.

  • Neurally Inspired Algorithms for Machine Vision and Learning
    Considerable progress has been made in the last three decades in designing efficient algorithms for specific applications in computer vision and machine learning. However, we still appear to be quite far from building intelligent systems that can exhibit the wide range of robust and adaptive visual behaviors that we as humans perform everyday. A promising approach to bridge this gap between human and machine vision is to exploit the emerging synergy between machine intelligence and computational neuroscience.

  • New Approaches to Identification Sensing
    Joshua Smith of Intel describes Electric Field Imaging, a non-contact means for tracking human body pose and motion that is now in broad commercial use for automobile passenger airbag suppression. This colloquia presentation also outlines FiberFingerprint, an authentication and identification technique for physical documents or objects that uses intrinsic, naturally occurring microscopic surface irregularities as a 'fingerprint.' Smith describes recent work with colleagues at Intel Research Seattle and University of Washington on WISP (Wireless Identification and Sensing Platform). WISP supports sensing, general purpose computation, and wireless communication, yet requires no batteries. WISPs are read and powered wirelessly, by ordinary standards-compliant Radio Frequency Identification (RFID) readers.

  • New Directions in Multiprocessor Synchronization
    Computer architecture is about to undergo, if not another revolution, then a vigorous shaking-up. The major chip manufacturers have, for the time being, simply given up trying to make processors run faster. Instead, they have recently started shipping "multicore" architectures, in which multiple processors (cores) communicate directly through shared hardware caches, providing increased concurrency instead of increased clock speed. As a result, system designers and software engineers can no longer rely on increasing clock speed to hide software bloat. Instead, they must somehow learn to make effective use of increasing parallelism. Transactional memory is a computational model in which threads synchronize by optimistic, lock-free transactions. This synchronization model promises to alleviate many of the problems associated with locking, and there is a growing community of researchers working on both software and hardware support for this approach. In this Distinguished Lecturer Series, Maurice Herlihy, Brown University, surveys the area, with a focus on open research problems.

  • Noun Phrase Coreference Algorithms
    Claire Cardie introduces noun phrase coreference resolution. She also describes the specific supervised and weakly supervised algorithms applied to the problem of noun phrase coreference resolution, presents empirical results on two standard coreference data sets, and discusses the problems encountered in applying each framework to the coreference task.

  • Nucleic Acid Logic Circuits for Programming Biology
    Learn about the design and experimental implementation of DNA-based logic gates and circuits in vitro. For their operation, the gates rely exclusively on sequence recognition and strand displacement reactions. Inputs and outputs are single-stranded nucleic acids and biological nucleic acids such as microRNAs can serve as inputs. The lecture will demonstrate logical AND, OR, and NOT, as well as thresholding and catalytic signal amplification.

  • Object Recognition Using Pictorial Structures
    This program considers the problem of recognizing objects from image data, using part-based models.

  • Object Recognition with Deformable Models
    The problem of detecting and localizing objects in images has important applications in a variety of areas, including robotics, image retrieval and medical image analysis. Deformable models represent objects as deformed versions of an ideal template. While this approach provides an elegant framework for object recognition, it also leads to difficult computational problems. The first part of this University of Washington program describes efficient algorithms that have been developed for finding objects in images using different types of deformable models. In the second part, Pedro Felzenszwalb of the University of Chicago considers the specific problem of detecting objects from generic categories such as people and cars in realistic scenes.

  • Object Recognition with Material and Shape
    One of the fundamental problems in computer vision is recognition. The two major sources of information for object recognition are material and shape. Material, commonly referred to as texture, is about what an object is made from. Shape, or geometry, is about what form an object takes. Presented are studies aimed at understanding different aspects of texture.

  • On the Power of Choosing the Shortest of Two
    Variations of a novel, recently proposed load balancing scheme based on small amounts of choice are investigated. Recently it has been shown that breaking ties in a fixed, asymmetric fashion, actually improves performance, whereas in all previous analyses, ties were broken randomly. The nature of this improvement using fluid limit models, suggest further improvements, and verify and quantify the improvement through simulations. Finally, practical applications of our improved hashing models are discussed.

  • Online Geospatial Data Sources
    With the explosive growth of the Web combined with improved technologies for remote sensing, there are now a huge number of geospatial data sources available online. This includes maps and satellite imagery. The challenge is how to make effective use of all this information. In this talk, Dr. Craig Knoblock describes integrating the huge amount of Web data with the widely available geospatial data sources.

  • Online Science: The World-Wide Telescope
    Jim Gray explains that computational science has historically meant simulation and why there is an increasing role for analysis and mining of online scientific data. As a case in point, half of the world's astronomy data is public. The astronomy community is putting all that data on the Internet so that the Internet becomes the world's best telescope: it has the whole sky, in many spectra, and in detail as good as the best 2-year-old telescopes. It is useable by all astronomers everywhere. This is the vision of the virtual observatory -- also called the World Wide Telescope (WWT). As one step along that path, Gray has been working with the Sloan Digital Sky Survey and CalTech to federate their data in web services on the Internet, and to make it easy to ask questions of the database. Gray explains the rationale for the WWT, and describes some the computer science challenges of publishing, federating, and mining scientific data.

  • Open Constraint Programming
    A framework for reactive programming over a shared store is presented. A key requirement is that the store can be modeled as a highly-structured entity such as a database or logical theory. The main elements are obtained from combining ideas in active databases, concurrent programming and constraints.

  • Open Data Kit: Open Source Mobile Data Collection
    Get ahead of the technology curve with Gaetano Borriello, Computer Science and Engineering professor at the University of Washington. Borriello reveals the latest data collection tool, Open Data Kit, a customizable mobile device that its creators hope will bring current research directly into the hands of citizen scientists, and public health and environmental communities. Supported by Google, ODK is sure to push information horizons in new directions.

  • Opening Doors: Mentoring on the Internet
    Students develop supportive relationships with adult mentors on the Internet. This video features participants in the DO-IT program at the University of Washington.

  • Operation-Centric Hardware Description and Synthesis
    High-level functional descriptions of hardware are often operation-centric in nature. An operation-centric description specifies the behavior of a system on an operation-by-operation basis.

    An operation-centric architectural description language based on the formalism of Term Rewriting Systems is presented. This synthesizable TRS language can capture complex hardware behavior concisely and precisely.

  • Our Infrastructures - Online And Vulnerable? Part 1 of 3
    From electricity usage to municipal water consumption, from traffic lights to dams, our world is monitored by computer systems. But the control systems that warn of impending problems may themselves be vulnerable to failure or attack because they are often online or otherwise hackable. Are they safe enough for lives to depend on them? Join a panel of experts who will take us into this invisible but critical world.

    This program is presented by The Center for Information Assurance and Cybersecurity at the University of Washington Information School and sponsored by UW Institute for National Security Education and Research and The UW Master of Strategic Planning for Critical Infrastructures online graduate program, with additional support fro The Information School, The Pacific Northwest Center for Global Security, and Pacific Northwest National Laboratory.



  • Our Infrastructures - Online and Vulnerable? Part 2 of 3
    This program includes a presentation by Mark Hadley, a research scientist with Pacific Northwest National Laboratory focusing on cybersecurity and the protection of critical infrastructure systems. He outlines the historical reasons for the vulnerability of networked digital control systems, and the technical requirements for better securing them. It also includes a presentation by Kevin Desouza, Ph.D., assistant professor in the Information School at the University of Washington and the director of the Institute for National Security Education and Research, an inter-disciplinary university-wide initiative. He speaks about the intelligence perspective on critical infrastructure protection, and the need for effective frameworks governing the gathering and use of intelligence.

    This program is presented by The Center for Information Assurance and Cybersecurity at the University of Washington Information School and sponsored by UW Institute for National Security Education and Research and The UW Master of Strategic Planning for Critical Infrastructures online graduate program, with additional support fro The Information School, The Pacific Northwest Center for Global Security, and Pacific Northwest National Laboratory.



  • Our Infrastructures - Online and Vulnerable? Part 3 of 3
    Part Three is a presentation by Dan Ryan, J.D., professor of systems management at National Defense University and an expert on information assurance and cyberlaw. He spoke about the highly varied legal issues -- many of them new -- that could flow from attacks on critical infrastructures via networked control systems, and how the existing legal system might or might not respond to those issues.

    This program is presented by The Center for Information Assurance and Cybersecurity at the University of Washington Information School and sponsored by UW Institute for National Security Education and Research and The UW Master of Strategic Planning for Critical Infrastructures online graduate program, with additional support fro The Information School, The Pacific Northwest Center for Global Security, and Pacific Northwest National Laboratory.



  • Our Job is EasyLiving
    The goal of our EasyLiving project at Microsoft Research in Redmond is to give users casual access to computing. Envisioned are rooms outfitted with cameras, microphones, displays, speakers, computers, and other devices that work together to move computing beyond the desktop into everyday life.

  • Overcoming Security Challenges in Emerging Technologies
    Emerging technologies have the potential to greatly improve the quality of our lives. Without the appropriate checks and balances, however, these emerging technologies have the potential to also compromise our digital (and physical) security and privacy. A key goal of the UW CSE Computer Security Lab is to help us achieve the best of both worlds: the wonderful promises offered by the new technologies without the associated security and privacy risks. This talk will examine several strands of our research, focusing first on our recent work with wireless implantable medical devices.

  • Paleontology Unit, Dinosaur Dig, and Materials Science and Technology Program
    A second-grade teacher at Franklin Elementary and a high-school materials science and technology teacher in Washington discuss programs they have developed to engage their students in contextual teaching and learning inside the classroom.

  • Parallel and Asynchronous Programming with F#
    Don Syme of Microsoft Research, Cambridge, looks at programming with F# in this video from the University of Washington. F# is a succinct and expressive typed functional programming language in the context of a modern, applied software development environment (.NET), and Microsoft will be supporting F# as a first class language in Visual Studio 2010. Syme offers an overview of F#, as well as some general coding, and takes a deeper look at each of these contributions and why they matter.

  • Parallelizing Programs using Approximate Code
    Zilles presents two hardware/software techniques to exploit thread-parallel resources to improve sequential program performance. The first technique uses "helper threads" to avoid stalls (due to cache misses and branch mispredictions) that would be incurred by a single-threaded execution of the program. The second technique uses a "master processor" to compute future checkpoints of program state to achieve a (speculative) parallel execution of the program. Both techniques exploit a common theme: the use of approximate versions of code to generate accurate value predictions.

  • Part 1 - Privacy: Reconciling Reality
    Hear a discussion about new federal and state laws meant to protect our privacy. What are the laws designed to achieve? How do they impact the general public, researchers, system managers and private organizations? In this program by The Center for Information Assurance and Cybersecurity at the University of Washington, learn strategies and tools under development to enhance information privacy and protection.

  • Part 2 - At Odds: Victims Rights vs. Free Speech
    Examine the muddy crossroads where online anonymous free speech can morph into harmful, targeted crimes of hate in this program by The Center for Information Assurance and Cybersecurity at the University of Washington. State Attorney General Rob McKenna introduces this moderated discussion on how two tenets of the American consciousness –- a deeply held belief in the value of laissez-faire economics and in the first amendment –- collide with real-life victims of Internet crime.

  • Pathways in Computer Science
    See the diverse pathways students pursue after receiving Bachelors degrees in computer science or computer engineering. Computer scientists work in a broad range of interesting fields, and an engineering degree is terrific preparation for almost any imaginable future. See profiles of computational biology, improving forest firefighting techniques, making textbook graphics readable to blind students, working in India to connect rural communities through technology, and creating better prosthetic devices for people with disabilities.

  • Perspectives on the Information Industry
    In this special colloquia, Dr. Eric Schmidt, chief executive officer of Google in Redwood City, California discusses the information industry and Google's approach to innovation. Dr. Schmidt has a 25-year record of achievement as an Internet strategist, entrepreneur and developer of great technologies. Since taking the helm at Google, he has focused on building the corporate infrastructure needed to maintain Google's rapid growth as a company and on ensuring that quality remains high while product development cycle times are kept to a minimum. Previously, Dr. Schmidt was chief executive and chairman of software maker Novell, and before that was chief technology officer at Sun Microsystems.

  • Peter at DARPA
    Peter Lee, director of the new Transformational Convergence Technology Office at the Defense Advanced Research Projects Agency (DARPA), shares TCTO's mission, research areas and approaches to technology innovation. TCTO aims to develop research programs that enhance the nation's ability to anticipate strategic surprise. It does this by re-establishing basic research programs in a broad range of rapidly emerging computing-enabled technology areas such as social media, synthetic biology, high-performance computing and networking, as well as employing a diverse range of innovation strategies including broad community programs, competitions and challenges, and crowd sourcing.

  • Photo Tourism and Photosynth: UW CSE, Microsoft Research, and Microsoft Live Labs Create a Winner
    Photosynth is an amazing new technology, created through a unique collaboration between Microsoft and the University of Washington. The application will change the way you think about digital photos by taking a large collection of photos of a place or object, analyzing them for similarities, and displaying them in a reconstructed 3-dimensional space. The groundbreaking innovation began with Photo Tourism research by UW graduate student, Noah Snavely, UW faculty member Steve Seitz, and Rick Szeliski, Microsoft Research. The new Microsoft Live Labs organization -- dedicated to advancing the state-of-the-art in Internet products and technology -- embraced the Photo Tourism technology and combined it with complementary photo browsing technology created by the team at Seadragon, a Madrona-backed Seattle startup headed by Blaise Aguera y Arcas that had been acquired by Live Labs. A Photo Tourism research presentation, and the rollout of the Photosynth prototype by Live Labs, stole the show at the 2006 SIGGRAPH conference.

  • PlanetLab: Evolution vs. Intelligent Design in Global Network Infrastructure
    PlanetLab is a global platform for evaluating and deploying network services. It currently includes nearly 700 nodes, spanning over 335 sites and 35 countries, and hosts over 500 experimental services. In this Distinguished Lecturer Series, Larry Peterson, Princeton University, identifies the requirements PlanetLab addresses, presents the design principles that follow from them, and outlines the resulting PlanetLab architecture. He also briefly discusses some of the lessons learned about building large network systems.

  • Power Aware Page Allocation
    One of the major challenges of post-PC computing is the need to reduce energy consumption, thereby extending the battery lifetime for mobile devices. Memory is an unexplored, and particularly important, target for efforts to improve energy efficiency. New memory technology offers power management features with the ability to put individual DRAM chips in any one of several different power modes. This talk explores the interaction of page placement with static and dynamic hardware policies to exploit these emerging hardware features.

  • Powerset and Natural Language Search
    Central to the concept of natural language search is that users express queries in natural language with the system responses respecting the linguistic information in the query. Architecture centered on natural language, linguistic and lexical knowledge translates directly into improved capabilities and experiences for end users. This creates both challenges and opportunities. Powerset is a startup company that is tackling these challenges in an attempt to bring natural language search to the world.

  • Practical Analysis Tools for Large-Scale Software
    UC Berkeley's Manu Sridharan presents two techniques -- refinement-based pointer analysis and thin slicing -- that enable powerful new tools for debugging and understanding large-scale software. His group’s refinement-based pointer analysis is the first to compute precise answers in interactive time, allowing tools to handle previously inscrutable program behaviors interactively. Thin slicing is the first technique to give usable answers to code relevance questions, a long-standing challenge for analysis tools. Sridharan also describes new tools enabled by the two techniques.

  • Practical Byzantine Fault Tolerance
    The growing reliance of our society on computers demands that we provide systems with improved reliability, availability, and security. This talk describes BFT -- a new software Byzantine fault tolerance toolkit that addresses these issues. BFT can be used to build replicated systems that work correctly and remain available even when some of their replicas behave arbitrarily due to malicious attacks, software errors, or hardware failures.

  • Pre-Execution: Staying on the Performance Curve
    Emerging applications continue to demand geometric increases in performance. Constraints like power, manufacturing cost, and the limits of physics itself mean that this performance will have to be obtained primarily by efficient extraction and exploitation of parallelism, rather than raw frequency or brute force solutions that provide only diminishing returns. Pre-execution is a technique that aggressively but efficiently finds parallelism in sequential codes -- codes that cannot be explicitly parallelized.

  • Princeton ZebraNet Project
    The field of wireless sensor networks offers many interesting applications. In this talk, Margaret Martonosi discusses the design tradeoffs that arise when applying wireless peer-to-peer networking techniques in a mobile sensor network designed for wildlife tracking. The ZebraNet system includes custom tracking collars carried by animals under study across a large, wild area that deliver logged data back to researchers. The domain-centric protocols and energy tradeoffs studied for ZebraNet will have general applicability in other wireless and sensor applications.

  • Proactive Computing
    For over 35 years, Information Technology has been dominated by J.C.R. Licklider's powerful vision of Interactive Computing / Man-Machine Symbiosis. Although this "Human Centered" line of research has been (and continues to be) tremendously productive, Tennenhouse presents the case for opening a "second front" in the research struggle -- one that zeros in on the boundary between the physical world, which we live in, and the virtual cocoons, in which we shelter our cherished software.

  • Proactive Computing: A Progress Report
    David Tennenhouse presents elements of IntelÕs research on proactive computing—working toward environments in which networked computers proactively anticipate our needs and take action—and provides a progress report on computing innovations. Tennenhouse also identifies some larger challenges, and encourages others in the research community to begin tackling them.

  • Probabilistic Methods for Mobile Robot Navigation
    Probabilistic methods are well suited for dealing with the uncertainties involved in sensing and acting in the real world. Dieter focuses on the application of probabilistic methods to fundamental problems in mobile robotics.

  • Probabilistically Checkable Proofs
    Khot gives a survey of PCPs, known hardness results, and techniques used to build PCPs. Computing approximate solutions is a way to cope with NP-complete problems. A tool called Probabilistically Checkable Proofs (PCPs) is used to establish hardness of approximation results.

  • Processes without Partitions
    Matthew Flatt demonstrates how a unique software solution can give programmers flexibility in using process-like constructs for managing tasks within a program. The PLT Scheme virtual machine design enables the implementation of the DrScheme programming environment and the PLT web server.

  • Program Analysis Techniques for Pointers and Accessed Memory Regions
    Virtually all program transformations and safety checks require information about the memory locations accessed by the statements and procedures in the program. But many constructs in modern programming languages make it difficult to obtain this kind of information.

    Presented are two new static analysis techniques, a pointer analysis and a symbolic bounds analysis, that precisely characterize how the program accesses memory.

  • Program Analysis: Binary Decision Diagrams
    Binary decision diagrams (BDDs) are a data structure that can efficiently represent large relations and provide efficient set operations. BDDs have traditionally been used for model checking, formal verification, and optimizing circuit diagrams. In John Whaley’s research, BDDs are applied to the area of program analysis. In this colloquia, Whaley describes a scalable context-sensitive, inclusion-based pointer alias analysis for Java programs. His approach to context sensitivity is to create a clone of a method for every context of interest and run a context-insensitive algorithm over the expanded call graph to get context-sensitive results.

  • Programmable Self Assembly
    Modern embedded and robotic systems are increasingly required to be completely decentralized. To build distributed sensor networks, automated factories, or autonomous robot teams requires tools for analysis, design, and synthesis that address the intertwined problems of modeling, communication, computation, control and complexity. Eric Klavins describes programmable self assembly as a means to address these issues.

  • Programming Language Ideas Escape the Lab: A Declarative Data Description Language for Managing Ad Hoc Data
    XML. HTML. CSV. JPEG. MPEG. These data formats represent vast quantities of scientific, governmental and industrial data. In an ideal world, all data would be in such formats. In reality, vast amounts of data exist in ad hoc formats, which do not have readily available tools. In this talk from the University of Washington, Kathleen Fisher describes the PADS data description language created to address this problem.

  • Programming Language Infrastructures
    Norman Ramsey explains that to be truly reusable, a language infrastructure needs not only a compile-time interface but also a run-time interface. He illustrates several examples and focuses on the mechanisms that make it possible for you to choose the semantics and cost model you want.

  • Programming: Secure Execution
    The greatest threat to our modern information infrastructure is the remote exploitation of program vulnerabilities—31 out of 34 CERT advisories year-to-date are due to such exploitations. Saman Amarasinghe explains many ways in which malicious code, typically inserted into a computer system masquerading as data, can allow unauthorized access to a vulnerable privileged program. Buffer overflow, format string vulnerabilities and execution models for legitimate programs will be discussed.

  • Project Halo: A Digital Aristotle
    Project Halo in a multi-staged effort towards a digital Aristotle, an application that will encompass much of the world's scientific knowledge and be capable of answering novel questions and advanced problem solving. Vulcan sees two primary functions for the digital Aristotle: first, as a tutor capable of instructing students in the sciences and, second, as a research assistant with broad interdisciplinary skills to help scientists in their work.

  • PUMA 2: Bridging the Processor/Memory Gap
    Increasing processor clock speeds, along with microarchitectural innovation, have led to a tremendous gap between processor and memory performance. The PUMA2 (Proactively Uniform Memory Access Architecture) project targets designing memory systems that bridge this gap in the presence of arbitrarily-irregular but repetitive memory access patterns. Babak Falsafi presents a memory model and technique which ends the debate over whether multiprocessors must support relaxed memory models.

  • Putting Computer Power to Work in an Online Learning Environment
    This program is about the developing of a system to push the limits exploiting computation in online learning in three principal areas: student construction, structured communication, and assessment without tests. Working in a freshman seminar called "Pixels, Numbers, and Programs" students solve image processing problems using mathematical and computational tools.

  • Puzzle Outsourcing
    One proposed approach to deter denial of service attacks is to require clients to solve computational puzzles before connecting to a server. Unfortunately, standard puzzle schemes impose a higher cost on legitimate clients than on attackers because legitimate clients must often solve puzzles online while users are waiting, but attackers can solve puzzles offline using hijacked machines. Ed Felten outlines a new type of puzzle scheme that lowers costs for servers and for legitimate clients (but not for attackers). The scheme outsources puzzle creation to a robust external service called bastion.

  • QED: A Simplifier for Concurrent Programs
    Explore QED, a new approach for reasoning about concurrency, with Shaz Qadeer of Microsoft Research in this video from the University of Washington. Since reasoning about complicated thread interleavings is difficult, QED simply avoids reasoning about them! Instead, QED simplifies the concurrent program iteratively, eliminating concurrency in favor of nondeterminism, producing in the limit a nondeterministic sequential program. The simplification performed by QED is based on a simple rewriting calculus. Yet, it is surprisingly powerful; an appropriate combination of these rewrite rules can often simplify concurrent programs dramatically. Qadeer also looks at other applications of the theory behind QED, such as programmer-assisted parallelization and static concurrency unit testing. and static concurrency unit testing. This talk is based on joint work with Tayfun Elmas and Serdar Tasiran.

  • Quantizing Time
    Many systems must treat time as continuous rather than quantized. If time is continuous, what does "first come first served" really mean. Like horse races that require a "photo finish" some choices of which task to do first are far from easy.

  • Quantum Information Science: A Quantum Computer Revolution
    Quantum computing represents the synthesis of fundamental quantum physics with concepts from computer science. In this talk, Dave Bacon introduces quantum computing and discuss two of the biggest questions in quantum computing: how does a quantum computer acquire its exponential power and, how, exactly, can we jumpstart a revolution in building a quantum computer? The answers to these questions will draw insights from the vast knowledge base of theoretical and experimental physics, mathematics, engineering, and computer science.

  • Query Processing in Sensor Networks
    Many of the emerging applications for sensor networks are focused on data collection and monitoring in remote environments. Samuel Madden, UC-Berkeley, discusses how the difficulties of coding power-management, routing and storage features in volatile distributed environments can be overcome by providing users with a simple, declarative interface. Madden also summarizes the query processing features of TinyDB.

  • Reasoning with Cause and Effect
    This talk summarizes concepts, principles, and inference tools that are useful in modeling aspects of causal reasoning. The principles are based on structural-model semantics, in which modifiable functional relationships, representing autonomous physical processes are the fundamental building blocks.

  • Reflective Interfaces
    Today’s computer interfaces are one-size-fits-all, and hard to change. Computer users with little programming experience have very limited opportunities to customize an interface to their task and work habits. Surmounting these problems requires a fundamentally different approach to UI design. Interfaces must be reflective in the sense that they are represented declaratively, can reason about their interactions with a user, and can change themselves in order to improve these interactions or in response to a user's explicit request. In this colloquia, UW CSE professor, Dan Weld, reviews a decade of work on intelligent interfaces at the University of Washington, distilling the lessons we have learned.

  • Regulatory Elements in Microbial Genes
    This program showcases an innovative University of Washington student-faculty academic project. The goal of the project was to write software that starts from a single microbial gene of interest, finds a large collection of corresponding genes from multiple microbes, and uses this collection to identify evolutionarily conserved patterns in their DNA regulatory regions.

  • Relational Artifacts: From Virtual Pets to Digital Dolls
    Sherry Turkle explores identity effects of the computer presence associated with new technology: Computational toys and digital "pets" affect how children understand what is and is not alive, and what is special about being a person. How do we conceptualize the nature of our attachments to interactive robots, and how does this interaction affect people's thoughts about themselves and their sense of identity?

  • Rendering by Manifold Hopping
    Presented is a novel image-based rendering technique called manifold hopping. The technique provides users with perceptually continuous navigation by using only a small number of strategically sampled manifold mosaics or multiperspective panoramas. Manifold hopping has two modes of navigation: moving continuously along any manifold, and discretely between manifolds.

  • Rendering Translucent Materials
    Accurately simulating the scattering of light by materials is fundamental for realistic image synthesis. Henrik Wann Jensen of Stanford University presents two practical techniques for simulating subsurface scattering.

  • Research at Google
    At its core, Google's mission is to organize the world's information and make it universally accessible and useful. Alfred Z Spector, vice president of Research and Special Initiatives at Google, shares the Internet giant’s approach to research innovation in this talk at the University of Washington. Spector shares some of Google’s most promising advances in translation, speech and vision, and considers computer science’s greatest future challenges.

  • Research Challenges in Software Radio
    A software radio is a wireless communications device that performs all of its signal processing in portable, application level software. By simply running a different program, a software radio can become a cell phone, wireless LAN, cordless phones, garage door opener or walkie-talkie. The flexibility enabled by software radio also enables faster technology tracking, since standards upgrade are simply software downloads. Software radio is a multi-disciplinary systems effort, bringing together research in digital signal processing, communications, operating systems, software engineering and algorithms. This talk highlights the research challenges in each of these areas and then describes some features that would be useful in a software radio domain specific processor.

  • Research in Educational Technology: Expanding Educational Possibilities
    The goal of the Educational Technology Group is to enhance education through novel deployments of computing technology. This includes improving the classroom experience by creating tools that allow greater flexibility in presentation and promote interaction, as well as extending the reach of education through different mechanisms for distance education. This program surveys a collection of projects the group has undertaken, including Tutored Video Instruction, Classroom Presenter, ConferenceXP, and Digital Study Hall. Learn about the challenges and future directions, including integrating a heterogeneous collection of devices to support active learning in the classroom, enhancing Tutored Video Instruction with greater support for facilitation, and creating an interactive environment in distance learning. Of particular interest are international course deployments and applications of educational technology to improve education in the developing world.

  • Rethinking Interconnect for High-Performance Computer Systems
    Over the past several decades, the ongoing miniaturization of silicon circuits has completely transformed the power and performance space of VLSI systems. Continued performance improvements may prove elusive, however, as both gate speed scaling and Moore's Law scaling slow down. Exacerbating these trends is the worsening bottleneck of chip-to-chip communication, driven by the growing disparity in size and power between off-chip and on-chip wiring. Ronald Ho of Stanford University and Sun Microsystems shares how researchers at Sun Labs are addressing these problems by dramatically reducing the costs of off-chip communication, allowing them to tile together multiple chips to create large "virtually monolithic" integrated circuits. The basic communication technology is one formed via the close positioning of two chips. This so-called "Proximity Communication" can use either capacitive coupling or optical communication.

  • Rethinking Internet Traffic Management Using Optimization Theory
    Dr. Jennifer Rexford, computer science professor at Princeton University, shares fresh ideas on how to better manage Internet traffic. Recent innovations in optimization theory now make it possible to develop more flexible, efficient protocols to control Internet traffic that satisfy both users and network operators.

  • Rick Devenuti, CIO, Microsoft Corporation
    Rick Devenuti graduated from the University of Washington with a Degree in Accounting. He then accepted a position in an auditing department of Deloitte Haskins and Sells. Within two years he was focusing on Small Business Consulting and Microsoft was one of his clients. In 1987 he accepted a position with Microsoft as a Cost Accounting Manager and soon realized he was a better fit for the company than he could have ever imagined. His next position as Controller of the U.S. Sales and Marketing Division gave him greater insight into the inner workings of the senior level of the organization. His next role as the Director of Financial Analysis gave him the opportunity to start a new group and reinforced his "failure is not an option" philosophy. He then became the General Manager of Sales Operations followed by his role as the General Manager of North American Operations where he was closely involved with the Windows 95 launch. He was then promoted to the Vice President of Worldwide Operations where he directed all Microsoft activities involving product information, product release, manufacturing and distribution vendors, transactions, processes and systems. He is currently heading up the Information Technology organization as the company’s CIO.

  • Robotic Reasoning
    Robots in the real world cannot plan in isolation. They need to worry about external agents that can change their environment while they are not looking. Unfortunately, reasoning exactly about external agents often results in intractable planning problems. Using the running example of teams of robots searching an area for moving evaders, Geoffrey Gordon describes new tools for simplifying and abstracting multi-robot reasoning problems.

  • Robust Design of Arithmetic Units for Error Tolerant Applications
    Josephine Ammer discusses designing robust arithmetic units for new semiconductors and wireless communications. As semiconductor technology is scaled, process variation becomes an ever-increasing problem. Designs incur a larger penalty to guarantee operation at the worst-case process corner. Many applications (wireless communication, audio and video signal processing, graphics, data mining, etc.) can tolerate certain levels of errors. The key challenge is ensuring those errors cause small changes in the final system output.

  • Routers: The Click System
    Kohler describes the Click system, the support and optimization tools, routing applications, and the design principles that make Click a success. The system attacks monolithic, in-kernel implementation problems with a configurable component framework for router construction.

  • Russ Daggatt, Teledesic
    Teledesic Corporation is ready to make a big impact. In fact it will revolutionize the world's communications. Founded by Craig McCaw and Bill Gates and supported by Boeing, Teledesic plans to deliver wireless broadband digital communications everwhere in the world through a system of 288 low earth orbit satellites. It will be the Internet in the sky. The price tag is $9 billion. Teledesic's president is a dynamic, personable attorney named Russ Daggatt. He has been guiding this extraordinary effort through the morass of international communications agreements, the Federal Communications Commission, development of cutting edge satellites and launch systems,and high stakes international finance.

  • SAGE: Software for Algebra and Geometry Experimentation
    SAGE is a University of Washington project whose goal is to create an optimal, free, open source software environment for research and experimentation in algebra, geometry, number theory, cryptography, and related areas. William Stein, a professor with the University of Washington Department of Mathematics started SAGE in 2005 by combining together the very best of existing free software, including: Singular, PARI, GAP, Macaulay2, Maxima, gfan. Next, he created interfaces to non-free software : MAGMA, Maple, Mathematical. From that point, he began to fill in the gaps with new code. Now dozens of developers have joined Stein in working on filling those gaps and making SAGE a polished and high quality piece of free software.

  • Scalar Operand Networks for Parallel Microprocessors
    Which computer architecture makes best use of a square centimeter of silicon? This is the question that drives microprocessor economics. In this talk, Michael Taylor examines one approach to transforming conventional microprocessors into scalable, parallel VLSI systems. The approach focuses on distributing resources across an interesting class of network _ the scalar operand network. Taylor examines the properties and performance of these networks, including insights from the scalar operand network in the 180 nm VLSI implementation of the 16-issue MIT Raw microprocessor.

  • Scaling Routers Using Optics
    Routers built around a single-stage crossbar and a centralized scheduler do not scale, and don't provide the throughput guarantees that network operators need to make efficient use of their expensive long-haul links. In a joint project between optics and networking research groups, Nick McKeown has been exploring how optics can be used to scale capacity and reduce power in a router.

  • Seattle Grand Challenges Summit Conclusion, Part 6 of 6
    In part six of the six part series chronicling the NAE Seattle Grand Challenges Summit,John Markoff, Correspondent, The New York Times host a panel discussion recapping the highlights from the conferences presentations on "Engineering Better Medicines" and "Engineering the Tools of Scientific Discovery". Plus a final audience question and answer segment. The panel included Matt O'Donnell, Frank and Julie Jungers Dean of Engineering, University of Washington, Larry Smarr, Founding Director of the California Institute for Telecommunications and Information Technology, University of California, San Diego.

    Ed Lazowska, Bill and Melinda Gates Chair in Computer Science and Engineering, University of Washington; Director, University of Washington eScience Institute; Chair, Computing Community Consortium and Bonnie Dunbar, President and CEO, Museum of Flight.



  • Securing the Web with Decentralized Information Flow Control
    This talk discusses how to secure both today's web sites and tomorrow's web computing platforms with a new OS technique called Decentralized Information Flow Control (DIFC). A DIFC system tracks the flow of secret data as it is copied from file to file and communicated from process to process. DIFC provides better security than standard OSes because it allows developers to concentrate security-critical code in small, audit-friendly declassifiers, which remain small and contained even as the overall system balloons with new features.

  • Security Protocols for Broadcast Communications
    Mr.Perrig describes how he designed and built a suite of new efficient security protocol families to enable broadcast authentication (TESLA & BiBa), broadcast signature (HTSS & MESS), and key distribution for large dynamic groups (ELK). His protocols are efficient, scalable, and tolerate high packet loss. He also discusses an implementation on a secure sensor network using nodes with sharply limited resources (8-bit microprocessor, 8-K ROM, 512 bytes RAM, limited battery life).

  • Seeing Through the Clutter
    The problem of analysing visual motion, especially against dense background clutter, is challenging. Uncertainty in the positions of visually sensed features and ambiguity arising from the clutter, call for a probabilistic treatment. A promising approach is to attempt to describe the image formation process probabilistically, and then look for suitable inference engines to do the analysis. Statistical fitering and expectation maximisation have both been tried as inference engines, with some success.

  • Self-Defending Software: Collaborative Learning for Security
    This University of Washington talk addresses software monoculture, many computers running the same application, which offers benefits for system administrators and users. But, every copy of the application is vulnerable to the same security exploits. The work discussed here by Michael Ernst of MIT enables a monoculture, or application community, to automatically defend itself against previously unknown zero-day exploits by creating patches that defeat those exploits without affecting application functionality.

  • Semiconductor Industry, Integrated Solutions
    The semiconductor industry faces an environment of increasingly complex challenges in chip manufacturing, global development and manufacturing, and cost-sensitive markets. Agnes Kwan describes how Intel uses an integrated research and development model to address these issues, including a discussion on the role of factory automation in technology development.

  • Semiconductor-Organic Heterostructures
    Adina Scott of Purdue University explores recent interest into incorporating molecular monolayers into electronic devices for sensing, nanoelectronic, energy conversion and biological applications. Device properties can be modulated using surface chemistry, leading to flexible fabrication schemes and new system functionalities. Examine key challenges related to the fabrication, structural characterization and electrical properties of such systems. Look at results for metal-molecule-silicon devices in which the electronic transport is governed by the interplay between the molecular-electronic properties and silicon bandstructure, enabling novel hybrid organic/semiconductor functionality. Combining these device concepts with optically-, biologically- or electrically-active molecular layers will result in new classes of hybrid devices with the potential for low-cost, highly-integrated systems.

  • Sharing and Abstraction in Hierarchical Reinforcement Learning
    One of the most interesting challenges for machine learning is learning for sequential decision-making (also known as Reinforcement Learning or RL). Tom Dietterich of Oregon State University describes the approach of Hierarchical Reinforcement Learning using the MAXQ value function decomposion, where the programmer defines a hierarchy of tasks and subtasks, and the value function is decomposed hierarchically into value functions for each task and subtask.

  • Side Channels and Clouds: New Challenges in Cryptography
    Emerging trends in computation such as cloud computing, virtualization and trusted computing require that computation be carried out in remote and hostile environments, where attackers have unprecedented access to the devices, the data and the programs. This poses new challenges for cryptography. Vinod Vaikuntanathan of the MIT/IBM T.J. Watson Research Center recently spoke at the University of Washington, sharing his recent work in solving two of these new challenges, side-channel attacks and computing on encrypted data.

  • Signal-Processing Framework for Forward and Inverse Rendering
    Understanding the nature of reflection and illumination is important in many areas of computer graphics and computer vision. In this talk, Ravi Ramamoorthi of Stanford University describes a new way of looking at reflection on a curved surface.

  • Simple, Scalable Network Algorithms
    The main idea of randomized algorithms is simple to state: rather than contend with a large state space, the trick is to make decisions based upon a few randomly chosen samples. Therefore, they lead to simple implementations of otherwise complicated algorithms. Balaji Prabhakar illustrates the use of randomization in devising high-performance network algorithms, specifically for switch scheduling, web caching, and bandwidth partitioning.

  • Smart Tools
    This short provides a brief view of the Smart Tools Academy, demonstratitng how it is helping steer Washington state toward full integration of technology in the classroom.

  • SmartBridge: A Scalable Bridge Architecture
    As the number of hosts attached to a network increases beyond what can be connected by a single local area network (LAN), forwarding packets between hosts on different LANs becomes an issue. Two common solutions to the forwarding problem are IP routing and spanning tree bridging.

    This talk describes a new architecture, called SmartBridge, that combines the good features of IP routing and spanning tree bridging.

  • Social Networks: Mathematical Models
    A social network, the graph of relationships and interactions within a group of individuals, plays a fundamental role as a medium for the spread of information, ideas, and influence among its members. In this talk, David Kempe investigates several optimization problems using the mathematical models of influence studied by sociologists.

  • Software and Systems Research
    Alfred Spector describes major components of IBM's research agenda and discusses aspects of IBM’s work in autonomic computing, such as the breakthroughs in unstructured information management with the "Blue Gene" project.

  • Software Synthesis for Embedded Systems
    The plummeting cost of hardware has enabled embedded systems to enter consumer products such as cellular phones and digital cameras. The cost of an error in such a high-volume device is skyrocketing, yet market pressures are forcing them to be developed ever more quickly. Despite this, many are still crafted with assembly language. Clearly, new approaches are needed.

  • Software Transactions: A Programming-Languages Perspective
    With multicore processors bringing parallel computing to the masses, there is an urgent need to make concurrent programming easier. Software transactions hold great promise for simplifying shared-memory concurrency, and they have received enormous attention from the research community in the last couple years. This talk will provide an overview of work done at the University of Washington to help bring transactions to the next generation of programming languages.

  • Some Hotter Topics in Information and Physical Security
    The entire technology world – not just IT – is constantly changing. Recent changes force a reexamination of assumptions about information security made only a brief time ago. Emerging technologies, once emerged, turn out to be a bit scary. Perhaps that's because the law of unintended consequences operates strongly at the bleeding edge, where tools end up cutting both ways. Many talk about "grand challenges in information security", while foolishly implementing things that are supposed to represent data but somehow end up being executable. At the moment, the most sophisticated measurements of security are things like bug counts, and do not address tradeoffs in managing security processes. Attacks which were thought to be impractical or uneconomical are now possible as a result of everything becoming smaller, faster, more sensitive, always connected, mobile, ubiquitous and cheap. Seiden’s talk lays out the landscape using stories and examples, and try to persuade you of the benefits of rethinking some of those assumptions to make your systems a bit more future-resistant.

  • Spectral Analysis for Data Mining
    Experimental evidence suggests that spectral techniques are valuable for a wide range of applications. A partial list of such applications include (i) semantic analysis of documents used to cluster documents into areas of interest, (ii) collaborative filtering --- the reconstruction of missing data items, and (iii) determining the relative importance of documents based on citation/link structure. Intuitive arguments can explain some of the phenomena that has been observed but little theoretical study has been done.

    A model for framing data mining tasks and a unified approach to solving the resulting data mining problems using spectral analysis is presented.

  • Spectral Clustering as Optimization
    Marina Meila analyzes why spectral clustering works, demonstrating that spectral algorithms work in a wider and more interesting range of cases than generally believed. In joint work with Jianbo Shi, Deepak Verma, and Liang Xu, Meila demonstrates that several popular clustering algorithms are equivalent near-perfect points. Meila proposes new methods for selecting the number of clusters and shows their superior performance in experiments.

  • Static Analysis of Dynamic Data Structures
    Dynamic allocation and destructive heap updates are fundamental language constructs that allow programmers to implement complex, efficient linked data structures. However, their flexibility makes it difficult for compilers and program analyzers to statically reason about the correct manipulation of such structures. Professor Radu Rugina discusses new heap analysis techniques and their application to error detection, program verification, and compiler transformations. These analyses are based on a novel approach where the compiler uses local reasoning about single heap cells, instead of global reasoning about the entire heap. This approach makes analyses precise enough to handle a large class of heap manipulation algorithms, and lightweight enough to scale to larger programs.

  • Statistical Failure Diagnosis in Software and Systems
    Alice Zheng presents a case study illustrating how statistical machine learning algorithms, along with appropriate system instrumentation, can aid in failure diagnosis. She proposes a statistical software debugging framework that collects information from past successes and failures via fine-grained instrumentation of the program and then analyzes this information to locate suspicious program predicates. Zheng discusses the algorithmic challenges of the approach and demonstrates a bi-clustering algorithm that is effective at simultaneously clustering failed runs and selecting useful predicates.

  • Statistical Knowledge Zero
    Zero-knowledge proofs, introduced by Goldwasser, Micali, and Rackoff, are fascinating constructs in which one party (the "prover") convinces another party (the "verifier") that some assertion is true, without revealing anything else to the verifier. Zero-knowledge proofs are powerful tools for constructing secure cryptographic protocols, and also yield rich classes of computational problems that are of complexity-theoretic interest as well. We investigate statistical zero-knowledge proofs, which are zero-knowledge proofs where the condition that nothing is revealed to the verifier is interpreted in a strong information-theoretic sense.

  • Statistical Learning Algorithms
    Michael Jordan discusses two broad classes of statistical learning algorithms. One is probabilistic graphical models, a formalism that exploits the conjoined talents of graph theory and probability theory to build complex models out of simpler pieces, and the second is kernel methods, a formalism based on computationally-efficient representations of generalized notions of similarity in Hilbert spaces.

  • Statistical Modeling: Genome Data
    Datasets, spanning many organisms and data types, are rapidly being produced, creating new opportunities for understanding the molecular mechanisms underlying human disease, and for studying complex biological processes on a global scale. Transforming these immense amounts of data into biological information is a challenging task. Eran Segal describes DNA patterns, regulatory modules, and gene expression profiles.

  • Stochastic Optimal Control in Biology and Engineering
    Control under uncertainty is a fundamental problem relevant to biology as well as engineering. Optimality models have explained numerous details of biological movements. Indeed optimal control and optimal (i.e. Bayesian) estimation are becoming the framework of choice for studying sensorimotor function. However most demonstrations of optimality are limited to relatively simple behaviors. In more complex and interesting behaviors we still lack the algorithms to compute what is optimal. Continued progress requires more efficient algorithms for stochastic optimal control. In this University of Washington program, Emanuel Todorov, of MIT and UCSD, presents a new problem formulation that greatly simplifies the construction of optimal control laws, and yields original algorithms.

  • Stream Programming: Luring Programmers into the Multicore Era
    As the computer industry has moved to multicore processors, the historic trend of exponential performance improvements will now depend on ordinary programmers and their ability to parallelize their code. However, most programmers are already overwhelmed by the complexity of modern software and are unwilling to expend extra effort on parallelization. Hence, for programmers to embrace a parallel abstraction, it must come with new capabilities that simplify application development and lure programmers into changing their ways. Learn more in this University of Washington program presented by William Thies of MIT.

  • STRONGMAN: Scalable Security Policy Management for Large Networks
    The design principle of restricting local autonomy only where necessary for global robustness has led to a scalable Internet. Unfortunately, this scalability and capacity for distributed control has not been achieved in the mechanisms for specifying and enforcing security policies. With the increasing size and complexity of networks, security management is becoming a more serious problem. This talk presents our architectural solution to this problem, STRONGMAN.

  • SUDS: Thread Level Speculation with Minimal Hardware Support
    In this presentation, Matthew Frank of MIT argues that compiler technology makes it possible for processors to parallelize a broader class of applications than is commonly believed.

  • Surface Computing and Computer Vision-Based Human Computer Interaction
    The vision of ubiquitous computing suggests that interactivity will be embedded throughout our physical environment in a wide variety of form factors and modes of use. Andy Wilson, a Microsoft researcher presents a series of projects which exploit sensing technologies such as computer vision to enable a wide variety of fluid, natural interactions situated on walls and tabletop surfaces. For example, PlayAnywhere is a compact tabletop projection-vision system which explores a number of new interactions on everyday surfaces, while TouchLight combines a transparent projection screen material with computer vision techniques. These new form factors have the potential of changing the way we relate to computing, but they also pose a challenge in terms of interaction design because they are so different from today's desktop computing.

  • System Errors and User Corrections in Spoken Dialogue Systems
    Understanding how people speak when they interact with spoken dialogue systems is critical to improving the performance of those systems. This talk describes several experiments on data collected from subjects using TOOT, a system for accessing train information by phone, showing a) that user corrections of system errors differ from other user speech along several prosodic dimensions; and b) that prosodic information, together with other automatically available features, can be used to classify turns as corrections with a high degree of accuracy.

  • Systematic Experimentation: When Abstraction Fails
    In this talk, Andreas Zeller illustrates how systematic experimentation automatically reveals the causes of program failures _ in the input, in the program state, or in the program code. Reasoning about programs is mostly deduction: the reasoning from the abstract model to the concrete run. Deduction is useful because it allows us to predict properties of future runs, up to the point that a program will never fail its specification. However, even such a 100 percent correct program may still show a problem: the specification itself may be problematic, or deduction required us to abstract away some relevant property. To handle such problems, deduction is not the right answer, especially in a world where programs reach a complexity that makes them indistinguishable from natural phenomena. Instead, we should enrich our portfolio by methods proven in natural sciences, such as observation, induction, and in particular experimentation.

  • Systems without Cooperation
    The Internet is no longer the cooperative, technological playground it once was. Successful networked systems must now account for potentially competing interests. In his lecture at the University of Washington, Dave Levin of the University of Maryland shares his research in how to keep participants in a networked system from deviating from the letter and spirit of a protocol. Doing so allows developers to create new mechanisms that foster cooperation among the otherwise self-interested.

  • Talking and Play
    Playtime is learning time. While children are playing theyre learning how the world works. You can help a child while youre playing together by engaging them in conversation about what theyre doing. While having fun, the child will be encouraged to communicate their thoughts by using new words.

  • Tangible User Interfaces
    Tangible user interfaces (TUIs) augment the physical world by integrating digital information with everyday physical objects. Developing tangible interfaces is challenging because programmers are responsible for acquiring and abstracting physical input. Scott Clemmer discusses the creation of Papier-Mâché, a toolkit for building tangible interfaces using computer vision, electronic tags, and barcodes. Papier-Mâché introduces high-level abstractions for working with these technologies that facilitate technology portability.

  • Technology for Developing Regions
    Moore's Law and the wave of technologies it enabled have led to tremendous productivity improvements and enriched the quality of life in the First World. Yet, technology has had almost no effect on the four billion people that make less than a dollar per day. In this talk, UC Berkeley researcher, Eric Brewer, argues that the decreasing cost of computing and wireless networking has created a unique opportunity: extending the benefits of technology to developing regions. Brewer illustrates software applications that have a very high impact and examines the research agenda for developing regions and their inherent economic and social issues.

  • Technology for Long-Term Care: Scaling Elder Care to the Next Billion
    Long-term care helps the elderly perform key day-to-day tasks such as eating, personal care and medication. Today, such care is overwhelmingly manual. However, the cost of manual care is unsustainable in the face of demographic trends. Without dramatic breakthroughs in the cost of care, over half of all elders are expected to be without adequate care within a generation.

    This talk describes a series of studies performed at Intel (in collaboration with several major partner organizations including the University of Washington) over the past six years towards understanding how technology may substantially reduce the manual burden of care.



  • Text Editing: Outlier Finding
    This talk describes how multiple selections can be used to automate repetitive text editing. A multiple selection is inferred from positive and negative examples given by the user. The selection is then used for typing, deleting, copying, pasting, or other editing. Multiple selection editing has been evaluated by user studies and shown to be fast and usable by novices. "Outlier finding" is a new way to reduce errors by drawing the user's attention to inconsistent data that may indicate errors. Miller has developed an outlier finder for text that can suggest both false positives and false negatives in a multiple selection. When integrated into the multiple-selection editor and tested in a user study, outlier finding reduced errors.

  • Text Mining with Information Extraction
    Information extraction (IE) locates specific pieces of data in natural language documents. Current research focuses on a form of text mining that first extracts a database from a document corpus using an IE system and then mines this database for interesting patterns using rule induction. Ray Mooney presents this discussion which includes automatically cleaning noisy textual databases and new rule-mining algorithyms.

  • The 2000 Biomedical & Health Informatics Symposium: A Vision for Informatics
    The symposium and panel discussion raise the fundamental theorum of medical informatics. The keynote speaker and panelists discuss how information is used, organized and retrieved with emerging technology in health-care, research and education. The symposium presents how the informatics approach carries an explicit focus on the interaction between the person and the machine as partners, which contrasts earlier technology approaches that focused primarily on the machine, and secondarily on the user.

  • The AltaVista Indexing and Search Engine
    An overview of how a web search engine is organized is provided. A key component of the AltaVista search engine: its indexing library, is described in more depth. The library manages a set of inverted files, and provides mechanisms to construct and optimize complex queries on those inverted files. The design goals were to enable efficient queries on bodies of text up to a few hundred gigabytes in size (e.g. AltaVista) without sacrificing too much generality, and without giving up on small applications (e.g. mail directories).

  • The Application of Platform-Based Design to Embedded Electronics and Synthetic Biological Systems
    Platform-Based Design is a design methodology within Computer Aided Design which at its core promotes the separation of functionality from implementation. Rigorous and formal applications of PBD have been shown to be very useful in the design of embedded electronic systems. This work has manifested itself in the development of the Polis, Metropolis, and Metro II design environments at UC Berkeley. PBD's true power lies in its ability to cross into new application areas. This talk will outline PBD techniques as they relate to both embedded electronics and synthetic biology.

  • The Computational Camera
    In this distinguished lecturer colloquia, Shree Nayar introduces the concept of a computational camera, a device that embodies the convergence of the camera and the computer, using new optics to select rays from the scene in unusual ways and an appropriate algorithm to process the selected rays. Nayar presents examples that demonstrate how the computational camera redefines the notion of an image and has the potential to impact the very nature of visual communication.

  • The Cyberspace Data Explosion: Boon or Black Hole?
    We are entering a cyber world where millions of sensors continuously collect data. From the ocean bottom to deep space, scientists are monitoring environments at unprecedented scales. On a more personal level, implanted medical devices can now monitor our well-being and "smart chips" embedded in passports, IDs and transit cards can track our comings and goings. Massive, ubiquitous databanks offer promise of great benefits but also dangers. How do we manage this data onslaught wisely? How do we guard our privacy and ensure our safety? UW scientists are asking these questions and blazing trails on the latest frontiers of cybersecurity.

  • The Design and Analysis of Simple Algorithms
    Allan Borodin, University of Toronto, addresses the question, "Is it time to make algorithm design more of a computer science?" A basic course in Computer Science undergraduate and graduate programs is "The Design and Analysis of Algorithms" or "Introduction to Algorithms." Looking at such course descriptions, syllabus, and textbooks, one can infer that an organizational theme is often in terms of "basic algorithmic paradigms." Maybe it is just that time does not permit us to illustrate many such paradigms or maybe there are not that many to illustrate. In any case, even though we usually "cover" very few such general techniques in our courses, we rarely if ever try to precisely define such algorithmic concepts and hence cannot rigorously address the question as to their power and limitations. Within the field of Operations Research there have been attempts to formally model and study dynamic programming and branch and bound algorithms but this work has been largely ignored in Computer Science.

  • The Desktop: Frontiers in Systems Research
    Desktop software, in the form of Web browsers, browser features and operating system distributions, are a growing area of engineering activity at Google. Brad Chen of Google, Inc. offers a look at Native Client as an example project in the space. Native Client is an open-source research technology for running x86 native code in Web applications, with the goal of maintaining the browser neutrality, OS portability and safety that people expect from Web apps. It supports performance-oriented features generally absent from Web application programming environments, such as thread support, instruction set extensions such as SSE and use of compiler intrinsics and hand-coded assembler. Google combines these properties in an open architecture designed to leverage existing Web standards and to encourage community review and third-party tools. Overall, Google's desktop efforts seek to enable new Web applications, improve end-user experience and enable a more flexible balance between client and server computing. Google has open sourced many desktop efforts, in part to encourage collaboration and independent innovation.

  • The End of Alchemy Empirical Software Security Assurance
    For more than a decade, scientists, visionaries and pundits have put forth a multitude of methodologies for building secure software, but what techniques deliver real results? Examine software security assurance as it is practiced today in this video from the University of Washington’s Computer Science and Engineering department. Hear in-depth interviews with leading enterprises such as Adobe, EMC, Google, Microsoft, QUALCOMM, Wells Fargo and Depository Trust Clearing Corporation. The lessons these leaders have learned can be applied in order to build a new effort from scratch or to expand the reach of existing security capabilities. Speakers also present a set of benchmarks for developing and growing an enterprise-wide software security initiative, including but not limited to integration into the software development lifecycle.

  • The End of Anonymity, the Beginning of Privacy
    The new Web economy relies on the collection of personal data on an ever-increasing scale. Information about our tastes, purchases, searches, browsing history, friendships and relationships, health history, genetics and more is shared with advertisers, marketers and researchers, raising a number of privacy issues. In this talk from the University of Washington, Vitaly Shmatikov, an associate professor of computer science at the University of Texas at Austin, considers several approaches to privacy-preserving data sharing and show that "anonymization," including popular methods based on k-anonymity and similar syntactic properties, fails to provide meaningful privacy guarantees.

  • The Exploration of Northwest Coast Indian Art
    The 27th Annual Faculty Lecture coincides with the series, "Contemporary Issues in Northwest Coast Native American Art," sponsored by the Burke Museum at the University of Washington. Using over 100 photographs of artworks, Professor Emeritus Bill Holm examines how Northwest Coast Native American Art has been perceived over the generations and what is going on today to understand it.

  • The Google Linux Cluster
    Google's Linux cluster currently processes over 150 million queries a day, searching a multi-terabyte web index for every query with an average response time of less than a quarter of a second, with near-100% uptime. In this discussion, Google Fellow Urs Hölzle will describe the software and hardware infrastructure that makes this performance possible, as well as provide an overview of the main problems facing a web search, software architecture, servers and compact rack hardware designs.

  • The Hidden Dimension of Shadows: A Rethinking of the Role of Shadows in Visualization
    Shadows have played a major role in visual arts but mainly as adjunct features that set a particular mood on a stage, help establish spatial order in a picture, or underscore the three dimensional aspects of solid forms. Recent work in sculpture however, has opened up a new potential for shadows: the role of carrying the narrative burden or "content" of the artwork. Professor Kagan shows examples of his steel and shadow sculptures that challenge a viewer's preconceptions of shadows, and discusses several new dimensional possibilities in them that may be useful for other applications.

  • The Impact of Multicore Architectures on Software: Disaster or Opportunity?
    How much information can you fit on a single microchip? Senior manager Michael Hind of the Programming Technologies Department at IBM Research reveals the most up-to-date changes in computer chip memory capacity, clock frequency and software optimization in this video from the University of Washington's Computer Science and Engineering Colloquium Series. Multiple processor cores on every chip will both challenge and revolutionize the industry.

  • The Institute for Systems Biology and Frontiers in Computational Biology
    Biology is now being viewed as an informational science. The Institute for Systems Biology emerged from the conviction that systems biology will be the major focus in biology and medicine in the 21st century and that cross-disciplinary partnerships (biologists, chemists, computer scientists, engineers, mathematicians, and physicists) are essential for realizing the potential of this new view. Lee Hood discusses the fundamental paradigm changes, in part, catalyzed by the Human Genome Project that lead to this view of systems biology.

  • The Intentional Domain Workbench
    The complexity of software code is the result of the intermingling of domain knowledge with implementation information. Generative Programming and Domain Specific Languages are known techniques for factoring and reducing the total complexity. The Domain Workbench that Intentional Software Corporation is developing makes the definition, creation, editing, combination, extension, and processing of DSL's more practical. Key features of the Domain Workbench are the uniform representation of multiple interrelated domains, the ability to project the domains in multiple editable notations, and direct access to the domain code by a program generator.

  • The Interdisciplinary Challenge of Building Virtual Worlds
    Creating an interactive experience requires talent from computer science, engineering, art, drama, design, architecture and a host of other disciplines. Randy Pausch presents his experience working with Walt Disney Imagineering on several virtual reality projects for the DisneyQuest "digital theme park."

  • The Limits of Quantum Computers
    In the popular imagination, quantum computers would be almost magical devices, able to "solve impossible problems in an instant" by trying exponentially many solutions in parallel. In this program, hear about four results in quantum computing theory that directly challenge this view.

  • The RFID Revolution
    Radio-Frequency Identification (RFID) is poised to replace bar-codes as the primary means of tracking and inventorying pallets and cases in the commercial supply chain. The uses of this technology, however, extend far beyond the supply chain, and RFID will soon become an integral part of our everyday lives. In this talk, CSE lecturer, Chris Diorio, will describe the fundamentals of RFID technology, the reasons for its impending adoption, the many benefits and looming issues, how it may impact the way we live and work, and the exciting research opportunities that await us.

  • The Stimson Bullitt Professorship in Environmental Law: Dedication of the Professorship
    The dedication of the Stimson Bullitt Professorship celebrates a new professorship at the University of Washington School of Law with a speech by Stimson Bullitt, Esq., and a multi-media presentation and lecture by Professor William H. Rodgers. Dean Roland Hjorth of the School of Law pays tribute to Mr. Bullitt, who has been active in the environmental movement in the Pacific Northwest since 1956.

  • The Structure of Information Networks
    The information we deal with is taking on an increasingly networked character. In part, this is because we now have the resources to represent and analyze enormous link-structured datasets. But networked content is also being created at a phenomenal rate; although the World Wide Web as we know it is less than a decade old, it is already being hailed as a new medium. These developments have led to an emerging study of complex networks, and a range of interesting research challenges.

  • The Theory of Secure Computation
    Conventional cryptography provides strongboxes for our secrets. However, information often must be combined with other information to be useful, and to do that we have to take it out of hiding. Or do we? For the last two decades, researchers have developed the field of secure computation, in which secrets are combined while minimizing the information leaked as much as possible. NEC Laboratories researcher Joe Kilian surveys the basics of secure computation and promising areas for future investigation.

  • The UrbanSim Project: Urban Simulation to Inform Public Decision-making
    The process of planning and constructing a new light rail system, expanding a freeway, or modifying zoning and land use plans is often politically charged. The goal in the UrbanSim project is to provide tools for planners, engaged citizens, and other stakeholders to be able to consider different scenarios, and then to evaluate these scenarios by modeling the resulting patterns of urban growth and redevelopment, of transportation usage, and of environmental impacts, over periods of 20- 30 years. Alan Borning, CSE, University of Washington, describes recent work on and applications of the project and gives some demonstrations.

  • Towards a Simpler Internet
    While the Internet has proven to be an astounding triumph of engineering, it continues to face huge technical problems. This talk will argue that many of these problems stem from the complexity of the Internet's protocol stack. Although the Internet is often easy to use, in the presence of faults, misconfigurations, attacks, and resource contention, the Internet's behavior can be very difficult to model or even understand. The research agenda at UW has been to develop a suite of protocols to radically simplify the Internet's observable behavior, to yield a more robust, securable, and efficient system. The lecture will give examples from interdomain routing, denial-of-service protection, congestion control, and wireless media access.

  • Towards Integrating the Sciences and Technologies of Learning for Education
    Dr. Pea's work involves trying to help create a seamless network between advances in technology, knowledge and learning. One of his aims it to create a national knowledge network for catalyzing best practices and new designs for improving learning with technologies among researchers, schools, and industries.

  • Traffic Matrix Inference and Anomaly Detection in Large IP Networks
    Albert Greenberg of Microsoft describes progress in IP network traffic matrix inference, arguably one of the most important technical problems in the engineering and management of large-scale IP networks. Greenberg discusses tomo-gravity (how to compute accurate traffic matrices for large ISPs in seconds), and anomography (how to accurately detect anomalies, at network-level, for large ISPs in seconds). The methods rely only on ubiquitously available link load and configuration data.

  • Training and Reality: Adventures of a Spaceflight Participant
    Space flight is still a very rare and exotic experience which has only recently been opened to "tourists", officially known as spaceflight participants. Charles Simonyi was the fifth of these as the 450th person in space. He will describe the decision process, the eight-month training and the flight itself from the point of view of a knowledgeable civilian, with particular emphasis on the issues of safety, traditions, and health aspects.

  • Transparent Interfaces: Letting the User in on the Secret
    A transparent interface reveals some of the inner workings of a system. Transparency is important in computer systems that must engender trust, that teach, or that must rely on users to correct their mistakes. Steve Tanimoto presents a framework for designing software systems to be transparent, and several methods are described for achieving important components of transparency. Examples are presented that illustrate transparency in (a) a system for teaching programming concepts, (b) image processing, and (c) online educational assessment.

  • Trends in Adaptive Computing
    The idea of changing the architecture of a processor at run-time to suit an application has been around for a long time. However, bringing adaptive (aka configurable) computing into the mainstream has been frustratingly difficult. This is unfortunate now that designers of high-performance embedded systems are searching for alternatives to ASIC’s (Application Specific IC’s), which are too inflexible and require high volumes, and processors like DSP’s, which cost too much and use too much power. Carl Eberling will first cover the history of configurable computing and describe some of the different research directions that have been taken. Ebeling will then present the research at the University of Washington to blend FPGA technology with processor architectures as an alternative to ASIC’s for high-performance, low-power computing for embedded applications.

  • Trends in Data Compression
    This program will describe some of the Data Compression Groups recent projects: (i) Group testing for image compression is a new flexible technique for achieving high quality lossy image compression and (ii) Unequal loss protection for compressed images and video is a way to obtain graceful degradation of fidelity in the presence of packet loss. It will also describe some of the basic principles in image and video compression including discrete cosine transforms, wavelet transforms, bit plane coding, and motion compensation. To conclude it will then describe several projects that have used group testing or unequal loss protection to achieve their results.

  • Ubiquitous Computing Research of the University of Washington
    This program discusses several ubiquitous computing applications that are being developed both at the University of Washington and Intel Seattle Research and use them to highlight the new research problems we face.

  • Ubiquitous Computing: System Privacy
    Hong discusses Context Fabric, a middleware architecture aimed at making it easier for developers to create privacy-sensitive context-aware applications. Context Fabric is based on Approximate Information Flows, a framework that Hong co-developed for analyzing privacy in ubicomp systems. Context Fabric provides an extensible suite of techniques both for application developers and for end-users.

  • UBIT Research Program: Technology for All
    Universal Benefit from Information Technology is a University of Washington program researching innovative approaches to provide access to, and use of, information technology. The research is especially relevant on a university campus which serves the needs of a diverse student population which needs access to desktop applications, web sites, and mobile devices. The u b i t research projects integrate information science, human-computer interaction, and computer science disciplines.

  • Uncommon Sense & Innovation
    Dr. William Brody, president of Johns Hopkins University presents his lecture, "Uncommon Sense and Innovation." Brody is the 13th president of Johns Hopkins University. With his extensive education in electrical engineering and medicine, Brody knows the importance of discovery and innovation in science. With that in mind, he discusses the role of reasoning and problem solving in the real world and emphasizes how to apply it to science.

  • Unconventional Vision Sensors
    What can be perceived by a human or computed by a machine from an image is fundamentally restricted by the captured data. Current imaging systems are severely limited in spatial resolution, field of view and dynamic range. Shree Nayar presents a discussion on new vision sensors that provide unconventional forms of visual information.

  • Undergraduate Research Symposium
    More than 150 University of Washington undergraduate students participating in research with faculty in 70 disciplines present their work at the University's second celebration of undergraduates in research.

  • Unifying Logical and Statistical AI
    Intelligent agents must be able to handle the complexity and uncertainty of the real world. Logical AI has focused mainly on the former and statistical AI on the latter. Markov logic combines the two by attaching weights to first-order formulas and viewing them as templates for features of Markov networks. Markov logic has been successfully applied to problems in information extraction, robot mapping, social networks and others, and is the basis of the open-source Alchemy system.

  • Unsupervised Learning
    There is precisely one complete language processing system to date: the human brain. Though there is debate on how much built-in bias human learners might have, we definitely acquire language in a primarily unsupervised fashion. Computational approaches to language processing, however, are almost exclusively supervised, relying on hand-labeled corpora for training. In this program, Dan Klein describes several syntactic representations designed to capture the basic character of natural language syntax as directly as possible. High-quality parses can be learned from surprisingly little text, with no labeled examples and no language-specific biases.

  • Usable Privacy and Security: Protecting People from Online Phishing Scams
    Phishing scams are a kind of semantic attack on computer systems that target the users of computer systems, rather than the system itself. Phishing is estimated to have cost over $3 billion in losses in 2008, with criminals impersonating banks, e-commerce sites, retail sites, and universities to trick people into giving them passwords or credit card information. Phishing attacks have also been used to steal sensitive information from corporations and governments, including the US Department of Defense. This talk will present ongoing research in protecting people from phishing attacks.

  • Using Evolution to Explore the Human Genome
    In this distinguished lecture, David Haussler talks about finding regions of the human genome that are not only under negative selection, but also are specifically evolving like protein-coding regions in genes. Haussler also investigates genetic innovations specific to primates and specific to humans. Given this as a base, and enough well-placed primate genomes to reconstruct intermediate states, scientists should eventually be able to document most of the genomic changes that occurred in the evolution of the human lineage from the mammalian ancestor over the last 80 million years.

  • UW Neural Systems Laboratory
    How does the brain parse sensory inputs to generate decisions and actions? Can signals measured from the brain be directly used to control prosthetics and other external devices? These are some of the questions driving research at the Neural Systems Laboratory at the University of Washington. CSE researcher Rajesh Rao discusses recent work on models of Bayesian inference in neurobiological circuits. Research in the lab is being pursued in collaboration with UW faculty in the Neurobiology and Behavior program, the UW Medical School, and the Institute for Learning and Brain Sciences.

  • Value-Sensitive Design: Informed Consent and Network Browser Security
    Many of us when we design and implement computer technologies focus on making the technology work -- reliably, efficiently, and correctly. Rarely do we focus on human values.

    Discussed are a collection of human values -- freedom from bias, autonomy, trust, informed consent, accountability, accessibility, access, and moral personhood -- that have been central to Friedman's research and design work.

  • Video Traces: Media Rich Annotations for Learning and Teaching
    Innovative uses of video, audio, text, and graphics make it possible for the learner to engage in knowledge creation activities and provide opportunities of participating in communities of practice. In this presentation, Reed Stevens show examples of how he has used rich media in his work and discusses his research which examines and compares cognitive activity in classrooms, workplaces, and science museums.

  • Virtual Private Machines: A Resource Abstraction for Multicore Computer Systems
    The computer industry is undergoing a momentous transformation. General-purpose computing is moving from desktops to diverse devices such as smart phones, digital entertainment centers and data center servers. At the same time, high-performance semiconductor manufacturers have shifted their focus from large monolithic processor designs to distributed multicore architectures. Learn more about architectures, multicore hardware, resource efficiency, and operating system policies in this University of Washington program with guest speaker Kyle Nesbit from the University of Wisconsin-Madison.

  • Visualmotor Tasks and Human Learning
    Submarine navigation and flight control tasks are difficult for humans to learn because they require the coordination of a task strategy (evasive maneuver) and the skills to implement that strategy (a visualmotor servo-loop). Under the Hybrid Learning Program of the Office of Naval Research, Devika Subramanian helped to create the NRL Navigation Task: design algorithms that learn probabilistic models of control policies used by humans from non-invasive recordings of their visualmotor activity. These models can be used by human or machine trainers to shape learning and also shed light on how humans acquire tasks with strategic and visualmotor components.

  • Voyagers and Voyeurs: Supporting Collaborative Information Visualization
    Interactive visualizations leverage human visual processing and cognition to increase the scale of information with which we can effectively work. However, most visualization research to date focuses on a single-user model, overlooking the social nature of visual media. Visualizations are used not only to explore and analyze, but to communicate findings. People may disagree on how to interpret data and contribute contextual knowledge that deepens understanding. Furthermore, some data sets are so large that thorough exploration by a single person is unlikely. Jeffrey Heer from the University of California, Berkeley, presents a number of novel visualization techniques in this University of Washington Computer Science and Engineering program.

  • Vulnerability Analysis
    Attack graphs represent the ways in which an adversary can exploit vulnerabilities to break into a system. In this talk, Jeanette Wing presents a technique, based on model checking, for generating attack graphs automatically. She also describes two kinds of analyses of attack graphs that system administrators can perform in trading off one security measure for another. This work is joint with Somesh Jha and Oleg Sheyner.

  • WaveScalar: Making Architecture Fun Again
    Silicon technology continues to provide an exponential increase in the availability of raw transistors. Effectively translating this resource into application performance, however, is proving increasingly difficult. Challenges such as wire-delay, design complexity, and manufacturing defects frustrate conventional approaches. WaveScalar, a new approach to building microprocessors, confronts these problems with a radical mix of architectural concepts. In this talk, Mark Oskin presents a broad overview of the WaveScalar research being conducted at the University of Washington.

  • Web Servers: The SYNC Project
    Research by Carnegie Mellon University’s Mor Harchol-Balter suggests the possibility of reducing the expected response time of every request at a Web server, simply by changing the order in which requests are scheduled. He proposes a method for improving the performance of Web servers servicing static HTTP requests by giving preference to requests for small files of requests with short remaining file size, in accordance with the SRPT (Shortest Remaining Processing Time) scheduling policy. The research proves both analytically and via implementation that SRPT scheduling can substantially improve mean response times over the traditional FAIR scheduling policy, without penalizing requests for large files.

  • What I Learned About Auctions Last Fall
    Last fall AT&T Labs participated in a project to help AT&T decide how to bid in the then-upcoming FCC spectrum auction, on which, in the end, AT&T spent approximately $2.9 billion. I will present some of the interesting and humorous facts and anecdotes I learned about real auctions, and why auction theory sometimes is, and sometimes isn't, applicable in the real world. This talk is definitely not intended for experts in auction theory!

  • Where Humans and Robots Connect
    A trailblazer in the emerging field of neurobotics, in 2007 Yoky Matsuoka received the MacArthur "genius" award. Now she's transforming our understanding of how the central nervous system coordinates musculoskeletal action. Her latest quest is to build the ultimate prosthetic: a fully functional replica of the human hand, controlled directly by the brain.

  • Whose Hat is That?
    Harvey quickly prepares the window display in time for a visit by Mr. Big.

    The undergraduate course on Computer Animation was taught to teams of students from Computer Science & Engineering, Architecture, Art, and Music by CSE Professor David Salesin, Cassidy Curtis from Pacific Data Images, and various guest lecturers.



  • WindowBox: A Simple Security Model for the Connected Desktop Colloquium
    Breaches in computer security do not just exploit bugs in applications; they are often also the result of mismanaged protection mechanisms. The tools available to protect sensitive resources and networks are tedious to use, non-intuitive, and often require expert knowledge. This paper presents a new security model, WindowBox, which presents the user with a model in which the workstation is divided into multiple desktops.

  • Women Entrepreneurs: Helen Rockey/Marjie Peterson
    Helen Rockey and Marjie Peterson detail how they built their companies and how women and men differ as business owners. Both Rockey and Peterson have led companies that have shown impressive growth. Rockey's company manufactures high quality running shoes that are sold worldwide. Peterson's firm supplies high tech employees to major Puget Sound area businesses.

  • Working with Digital Natives
    What makes someone a digital native? Are digital natives defined by their generation, or the technology they use? Explore this term, and the differences between those considered digital natives and those considered digital immigrants, as Michael Eisenberg, dean emeritus of the iSchool at the University of Washington, leads a diverse panel in an discussion of these topics.

  • World Wide Access: Accessible Web Design
    Creating an accessible website involves making your information accessible to all visitors, including those with disabilities. Learn the essential elements for designing a site that is perceivable, operable, understandable, and robust.
Support for UWTV is provided by:

Contact Us | Help | Where to Watch UWTV | Retransmission | UWTV Production | uw2.tv
Email us with comments or questions or call 888-616-UWTV

Copyright © University of Washington, 1997-2010. All Rights Reserved.