Colloquium Archive

September 13 Wednesday 1:30 - 2:30 369 Link Hall	Title: Recent Progress in Network Coding Speaker: Dr. Gerhard Kramer (Bell Laboratories) Abstract: With its inception at the turn of this century, network coding has become a rapidly evolving area in Network Information Theory. In this talk, we will examine the basic concept behind network coding and contrast it to the conventional approach in existing computer and communication networks. We highlight the capacity gain by allowing each node in a network to re-encode the input information instead of simply replicating and routing the information. We will then discuss some recent progress in network coding. A bound on network coding rates is presented that generalizes an edge-cut bound on routing rates. The bound, called a progressive d-separating edge set (or PdE) bound, involves progressively removing edges from a network graph and checking whether certain strengthened d-separation conditions are satisfied. We show that the PdE bound and one of its extensions proves that routing is rate-optimal for multiple unicast sessions on bidirectional ring networks. We further show that the PdE bound improves on a standard cut-set bound for networks with broadcasting, interference, and noise. Bio: Gerhard Kramer received the B.Sc. and M.Sc. degrees in electrical engineering from the University of Manitoba, Winnipeg, MB, Canada, in 1991 and 1992, respectively, and the Dr. sc. techn. (Doktor der Technischen Wissenschaften) degree from the Swiss Federal Institute of Technology (ETH), Zurich, Switzerland, in 1998. From July 1998 to March 2000, he was with Endora Tech AG, Basel, Switzerland, as a communications engineering consultant. Since May 2000 he has been with the Communications and Statistical Sciences Department (formerly the Mathematics of Communications Research Department), Bell Laboratories, Lucent Technologies, Murray Hill, NJ, USA. His research has been focused on information theory, communications theory, iterative decoding, and source coding.
September 27 Wednesday 1:30 - 2:30 369 Link Hall	Title: Discovering, Expressing, and Conforming to Privacy Rules Speaker: Dr. Carl A. Gunter (University of Illinois Urbana-Champaign) Abstract: This talk will overview research aimed at improving the ability to determine and express rules for privacy, and assure that systems dealing with private data conform to such rules. We will review three studies. The first study examines a framework for managing information generated by Building Automation Systems (BASs) to support Location Information Systems (LISs) that track building users. This approach is used to define a privacy-preserving LIS for a smart building. The second study demonstrates a formalism called "privacy APIs" for describing legislative rules precisely. We show how the formalism is able to encode the US HIPPA consent rules for patient records and how this encoding can use a model checker to examine key properties. The third study presents the results of a survey on privacy attitudes intended to guide privacy policy and aid the design of interfaces for allowing individuals to choose their own rules for privacy. The survey reveals attitudes about privacy in diverse sectors and according to the parameters of collection, access, and use. Bio: Dr. Gunter received his BA from the University of Chicago in 1979 and his Ph.D. from the University of Wisconsin at Madison in 1985. He worked as a postdoctoral researcher at Carnegie-Mellon University and the University of Cambridge in England before joining the faculty of the University of Pennsylvania in 1987. He joined the University of Illinois at Urbana-Champaign in 2004 where he is a professor, Director of the Illinois Security Lab, member of the Arms Control, Disarmament and International Security executive committee, and member of the Information Trust Institute Steering Committee. He does research and teaches at UIUC in his areas of technical expertise: security, networks, programming languages, and software engineering. His work includes contributions to the foundations of programming languages, the design of functional and object-oriented programs, languages and models for networks and security, and software engineering. He has published over 70 papers in scientific forums, advised 7 Ph.D. theses, and authored an MIT Press textbook on the semantics of programming languages. Dr. Gunter has acted as a technical consultant for a number of companies, including AT&T, Oki Electric, Lucent Technologies, and Intertrust. He has also provided legal advising and acted as an expert witness in the areas of fraud, contract, patent, and copyright. He has acted as a principal investigator on numerous research grants, including grants from the Army, Navy, DARPA, NSF, Cisco, Microsoft, NEC, and SAIC. In 2000, he founded Probaris Technologies, a security technology company in Philadelphia, where he served as director, chief scientific advisor, and software architect.. He is the author of a number of pending patents.
October 11 Wednesday 1:30 - 2:30	Title: Coin-Flipping for Satisfaction and Beyond Speaker: Dr. Carla Gomes (Cornell University) Abstract: Randomized search methods have greatly extended our ability to solve hard computational problems, In recent years, we have seen the emergence of an active research area focusing on the study and design of complete (exact) randomized search procedures. In this talk I describe a new generation of search methods that exploit randomization while guaranteeing completeness. I will show that the run time distributions of complete randomized search methods are characterized by a large variance in performance, exhibiting intriguing properties, often characterized by very long tails or ``heavy tails''. Such non-standard distributions have recently been observed in areas as diverse as economics, statistical physics, and geophysics. I will discuss how one can boost the performance of search procedures by exploiting the heavy tail phenomena, demonstrating speedups of several orders of magnitude for state-of-the-art complete search procedures running on hard, real-world problems. I will also discuss formal models of heavy-tails in combinatorial search and discuss different ``statistical regimes of heavy-tailed behavior'' in combinatorial domains, providing a general characterization of parameter regions where heavy-tailed behavior is prevalent. In addition I will introduce streamlining reasoning, a technique inspired by a real-world problem and motivated by the intriguing theoretical properties of random parity constraints, very effective for solution finding, counting, and sampling, dramatically boosting combinatorial search on real-world instances Bio: Carla Gomes is Associate Professor in the Faculty of Computing and Information Science, Department of Computer Science and Department of Applied Economics and Management at Cornell University. She is also the director of the Intelligent Information Systems Institute (IISI) at Cornell. Carla obtained her Ph.D. in computer science in the area of artificial intelligence and operations research from the University of Edinburgh in 1993. She also holds a M.Sc. in applied mathematics from the University of Lisbon. Her research has covered many areas in artificial intelligence and computer science, including planning and scheduling, integration of constraint and mathem atical programming techniques for solving combinatorial problems, complete randomized search methods, and algorithm portfolios. Carla's research spans the full range of theory to applications. Carla's current projects focus on the interplay between problem structure and computational hardness, the use of approximation methods in large scale constraint-based reasoning systems, and applications of constraint-based reasoning and optimization to combinatorial problems such as those arising in combinatorial design, autonomous distributed agents, and most recently, combinatorial auctions and science of networks. Carla was the program co-chair of the Ninth International Conference on Theory and Applications of Satisfiability Methods (SAT 2006) and conference chair of the Eighth International Conference on Principles and Practice of Constraint Programming (CP 2002). Recently, Carla received best paper awards at 10th Conference on the Principles and Practice of Constraint Programming, 2004, and at the 21st National Conference of the American Association for Artificial Intelligence (AAAI-06).
November 2 Thursday 1:30 - 2:30 347 Hinds Hall (Please note that the day and the place are different from our other colloquiums).	Title: Sensitive Information in a Networked World Speaker: Dr. Joan Feigenbaum (Yale University) Abstract: Increasing use of computers and networks in business, government, recreation, and almost all aspects of daily life has led to a proliferation of online sensitive data, i.e., data that, if used improperly, can harm the data subjects. As a result, concern about the ownership, control, privacy, and accuracy of these data has become a top priority. Since fall of 2003, the PORTIA project (http://crypto.stanford.edu/portia) has focused on both the technical challenges of handling sensitive data and the policy and legal issues facing data subjects, data owners, and data users. The PORTIA goals are (1) to design and develop a next generation of technology for handling sensitive information that is qualitatively better than the current generation's and (2) to create an effective conceptual framework for policy making and philosophical inquiry into the rights and responsibilities of data subjects, data owners, and data users. This talk will review some of the progress and conclusions of the project's first three years. About the Speaker: Joan Feigenbaum is the Henry Ford II Professor of Computer Science at Yale University. She received a BA in Mathematics from Harvard and a Ph.D. in Computer Science from Stanford. Between finishing her Ph.D. in 1986 and starting at Yale in 2000, she was with AT&T, where she participated very broadly in the company's Information-Sciences research agenda, e.g., by creating a research group in Algorithms and Distributed Data, of which she was the manager in 1998-99. Professor Feigenbaum's research interests include Internet algorithms, computational complexity, security and privacy, and digital copyright. While at Yale, she has been a principle in several high-profile activities, including the NSF-funded PORTIA Project and the ONR-funded SPYCE Project. Her current and recent professional service activities include membership in the NAS Computer Science and Telecommunications Board, Program Chair for the 2002 ACM Workshop on Digital Rights Management, Program Co-Chair for the 2004 ACM Conference on Electronic Commerce, Conference Chair for the 2006 ACM Conference on Electronic Commerce, Executive-Committee Member-at-Large of the ACM Special Interest Group on Algorithms and Computational Theory (Sigact), and Vice Chair of the ACM Special Interest Group on Electronic Commerce (Sigecom). Professor Feigenbaum is a Fellow of the ACM.
November 8 Wednesday 1:30 - 2:30 369 Link Hall	Title: Information Fusion and Machine Intelligence for Biological Sensing and Decision-Support Systems Speaker: Dr. Jerome J. Braun (Lincoln Laboratory, Massachusetts Institute of Technology) Abstract: Demanding decision-making tasks -- in biodefense, medical or industrial diagnostics, environmental engineering, and many other areas -- can often benefit from, and some of them require, reliance on multiple information sources of diverse modalities. In many situations, the exploitable disparate multisource data ^Ö from multiple sensors and other sources -- exhibit various forms of imperfection, such as uncertainty, inconsistencies, conflicts, incompleteness, ambiguity and vagueness. Automatic fusion and decision-making in these circumstances is challenging, especially when the domain phenomena cannot be modeled adequately because of an insufficient a priori knowledge. This presentation will argue for the role of machine learning and reasoning methods and certain forms of their integration as a means of addressing these issues and -- more broadly -- as a path to intelligent information-fusion for biological defense and other demanding decision-support applications. The discussion will be in the context of selected research efforts we have undertaken in recent years. The first two of the discussed efforts are aerosol anomaly detection for subway biodefense and microarray pattern recognition for pathogen identification. The third effort, currently in progress, addresses the problem of multisource information fusion for biological sensor networks with a novel cognitive-processing hybrid information-fusion architecture. The discussion will include an overview of that architecture and some of its constituent methods. While the specific application domain of the discussed efforts is that of bioattack detection, the approaches and methods developed are relevant to a wide range of other decision-making problems, and can bring us closer to the goal of intelligent decision-support systems. This material is based upon work sponsored by the Department of the Air Force under Air Force Contract #FA8721-05-C-0002. Opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the United States Government. This material is based upon work supported by the National Science Foundation under Grant No. 0329901. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation. About the Speaker: Dr. Jerome J. Braun holds a Full Staff position at Lincoln Laboratory, Massachusetts Institute of Technology (MIT). His current research is focused on information fusion and machine intelligence for biological defense. His earlier research at MIT Lincoln Laboratory was in the area of ballistic missile defense. Before joining MIT Lincoln Laboratory, Dr. Braun held a Principal Staff position at GTE Laboratories, where much of his research concentrated on advanced speech recognition systems. He also was an Adjunct Professor at the University of Massachusetts Lowell, teaching graduate courses in Automatic Speech Recognition. Prior to joining GTE Laboratories, his work included the areas of manufacturing processes automation, and advanced military avionics airborne systems for aircraft navigation and control. Dr. Braun holds a B.Sc. degree in Physics, a M.Sc. degree in Computer Science, and a Sc.D. degree in Computer Science. His publications include the areas of machine intelligence, multisensor multisource information fusion, bioinformatics, pattern recognition, signal processing, speech recognition, and language identification.
November 29 Wednesday 1:30 - 2:30 369 Link Hall **CANCELLED**	Title: Distributed Sensor Networks – A Perfect Example of Interdisciplinary Science and Engineering Speaker: Dr. S. S. Iyengar (Chairman at Louisiana State University) Abstract: Distributed Sensor networks have a wide range of real-time applications in aerospace, automation, defense, medical imaging, robotics, and weather prediction. Over the past several years, scientists, engineers, and researchers in a multitude of disciplines have been clamoring for more detailed information without much success. This new evolving technology can provide solutions to a variety of these technology related problems. Professor Iyengar in this talk will give an overview of these impact areas based on his experience in working with various industries like Oak ridge National Lab, Jet Propulsion Lab, and Naval Research Lab. His talk is also based on his experiences of working with scientists from Raytheon, Boeing during the last few years and giving workshops/talks at some of these industries. About the Speaker: He is the Chairman and Roy Paul Daniels Chaired Professor of Computer Science and is also Satish Dhawan Chaired Professor at the Indian Institute of Science. His publications include 13 books (authored or coauthored textbooks; Prentice-Hall, CRC Press, IEEE Computer Society Press, John Wiley & Sons, etc.) and over 300 research papers in refereed journals and conferences. He has served as a Guest Editor for the Journal of Theoretical Computer Science and the Journal of Computer and Electrical Engineering. He has been involved with research in high-performance algorithms, data structures, sensor fusion, data mining, and intelligent systems. He is also a Fellow of the IEEE, ACM and a Fellow of American Association of Advancement of Science (AAAS). He has won many best paper awards a member of European Academy of Science. Dr. Iyengar was awarded the Distinguished Alumnus Award at the Indian Institute of Science in March 2003. He has served as an Editor for the IEEE Transactions on Computers, the IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, the IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS, and the IEEE TRANSACTIONS ON SOFTWARE ENGINEERING, etc. He is also the founding editor of “The International Journal of Distributed Sensor Networks”. His funding comes from NSF, DARPA, ONR, US Army Research Office, Naval Research Laboratory, MURI, JPL/CalTech programs.

Colloquium Archive