Conferences, Lectures, & Seminars
Events for April
-
Bekey Keynote Speaker
Thu, Apr 05, 2007 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Ed LazowskaBill & Melinda Gates Chair
in Computer Science & Engineering University of WashingtonTitle: Computer Science: Past, Present, and FutureAbstract: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, and I'd like to take this opportunity to engage you. The next ten years of advances in computer science should be far more significant, and far more interesting, than the past ten. I will review the progress that our field has made, and I will present a number of "grand challenge" problems that we should be prepared to tackle in the coming decade. I'll invite your contributions.Biography:Ed Lazowska holds the Bill & Melinda Gates Chair in Computer Science & Engineering at the University of Washington, where he has been on the faculty since 1977. Lazowska's research and teaching concern the design, implementation, and analysis of high performance computing and communication systems. He is a Member of the National Academy of Engineering, and a Fellow of the American Academy of Arts & Sciences, ACM, IEEE, and AAAS. He has chaired the NSF CISE Advisory Committee, the DARPA Information Science and Technology (ISAT) Study Group, and the Computing Research Association Board of Directors. He has been an advisor to Microsoft Research since its inception in 1991, and serves as a board member or technical advisor to a number of high-tech companies and venture firms.Location: Ethel Percy Andrus Gerontology Center (GER) - Auditorium
Audiences: Everyone Is Invited
Contact: Nancy Levien
-
CS Colloquium- Even-Dar
Fri, Apr 06, 2007 @ 03:30 PM - 04:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Title:Limitations and Challenges in No Regret AlgorithmsEyal Even-DarUniversity of PennsylvaniaAbstract:No regret algorithms have been studied extensively since the pioneering works of Blackwell, Hannan and Robbins in 1950s. In a sequential decision making problem where you are given an advice by N experts at each time step, the no regret algorithms guarantee that you will always do almost as good as the best expert in hindsight. Equipped with such strong theoretical guarantee, no regret algorithms have been applied successfully in many fields including machine learning, economics and game theory.In this talk I will present several challenges to and limitations of no regret algorithms. More specifically, I will consider adding natural constraints and adapting no regret algorithms to dynamic environments. Such scenarios will shed a light on how far these algorithms can be "pushed", while still satisfying the no regret property. Finally, I will consider a routing game where all users use no regret algorithms, and study the efficiency and the stability of such game.Biography:Eyal Even-Dar received his Ph.D. from Tel Aviv University in
2005 and is currently a postdoctoral fellow at the University of Pennsylvania. His interests include computational learning theory, machine learning, game theory and their intersection.Hosted by David KempeRefreshments will be served.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Nancy Levien
-
CS Distinguished Lecture Series
Thu, Apr 12, 2007 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
This lecture has been cancelled.Dr. John GrayMicrosoft eScience GroupTitle: eScience -- bringing all the world's science data and literature online and cross-indexing it.Hosted by Prof. Shahram Ghandeharizadeh
Audiences: Everyone Is Invited
Contact: Nancy Levien
-
Parallel Multifrontal Sparse Solvers
Thu, Apr 12, 2007 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Robert LucasDirector of the Computational Sciences Division
USC Information Sciences InstituteAbstract:Solving large sparse systems of linear equations is the computational bottleneck in many applications. These include such diverse fields as mechanical computer aided engineering (MCAE) and interior point methods from optimization. Therefore, over the last three decades, a great deal of research has gone into porting such algorithms to large-scale, parallel systems. This talk will review the multifrontal method for factoring such linear systems and the experience gained porting it to distributed memory (MPI), shared memory (OpenMP), and SIMD systems. The talk will conclude with lessons learned and a discussion of some of the open research problems in this area.Biography:Dr. Robert F. Lucas is the Director of the Computational Sciences Division of the University of Southern California's Information Sciences Institute (ISI). There he manages research in computer architecture, VLSI, compilers and other software tools. Prior to joining ISI, he was the Head of the High Performance Computing Research Department in the National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory. There he oversaw work in scientific data management, visualization, numerical algorithms, and scientific applications. Prior to joining NERSC, Dr. Lucas was the Deputy Director of DARPA's Information Technology Office. He also served as DARPA's Program Manager for Scalable Computing Systems and Data-Intensive Computing. From 1988 to 1998 he was a member of the research staff of the Institute for Defense Analyses, Center for Computing Sciences. From 1979 to 1984 he was a member of the Technical Staff of the Hughes Aircraft Company. Dr. Lucas received his BS, MS, and PhD degrees in Electrical Engineering from Stanford University in 1980, 1983, and 1988 respectively.Host: Aiichiro Nakano
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Nancy Levien
-
CS Colloquium - Rubenstein
Thu, Apr 19, 2007 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Network Resilience: Improving Survivability, Security, and Robustness of Emerging Network SystemsDan RubensteinAssociate Professor of Electrical Engineering and Computer Science Columbia UniversityAbstract:Computer networks of today and tomorrow need to be deployed rapidly and operate in environments where there is limited cooperation and trust among the nodes composing the network. Examples of such networks include sensor network deployment in a disaster setting, competing 802.11 hotspots, cooperative mesh networks, and delay tolerant networks (DTNs). Such networks are inherently less predictable and more susceptible to accidental or intentional abuses. Our research at Columbia has focused on resilience for this more vulnerable space of networks: how to make them function properly and efficiently when the infrastructure is unplanned, untrusted, insecure, undergoes rapid change, or is attacked.I will begin by describing the various projects in resilience that our lab has focused on over the past several years, and then focus specifically on the problem of control plane monitoring for routing protocols. Distributed routing protocols traditionally assume that all nodes executing the protocol can be trusted to truthfully and correctly report control plane information, but history has demonstrated that sometimes inaccurate information can be propagated with devastating consequences. We develop a theoretical framework that allows us to understand when, using state information provided by a distributed routing protocol, this information can be used to detect erroneous propagation of information. We derive a polynomial-time algorithm for distance vector (Bellman-Ford) and Path-Vector (BGP) style protocols called Strong Detection and prove that if our algorithm cannot detect an error, then the error is undetectable, given the existing state information. We conclude by showing our ongoing work on applicability of Strong Detection to wireless ad-hoc network settings.Biography:Dan Rubenstein is an Associate Professor of Electrical Engineering and Computer Science at Columbia University. He received a B.S. degree in mathematics from M.I.T., an M.A. in math from UCLA, and a PhD in computer science from University of Massachusetts, Amherst. His research interests are in network technologies, applications, and performance analysis, with a substantial emphasis on resilient and secure networking, distributed communication algorithms, and overlay technologies. He has received an NSF CAREER Award, an IBM Faculty Award, the Best Student Paper award from the ACM SIGMETRICS 2000 conference, and a Best Paper award from the IEEE ICNP 2003 Conference.Hosted by Leana GolubchikRefreshments will be served.
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Nancy Levien
