_______________________________
Jointly Sponsored by Department of Electrical Engineering and Computer Science and CASE Center
http://www.lcs.syr.edu/academic/dept_electricalengcompsci/colloquium.aspx
Jointly Sponsored by:
Department of Electrical Engineering & Computer Science and the CASE Center

Colloquium Archive

 

September 16               Wednesday 1:00PM                  4-201 Center for Science and Technology   Dr. John C. Reynolds	Title: Toward A Grainless Semantics For Shared-Variable Concurrency Speaker:  Professor John C. Reynolds, Carnegie Mellon University Abstract: Conventional semantics for shared-variable concurrency suffers from the "grain of time" problem, i.e., the necessity of specifying a default level of atomicity. We propose a semantics that avoids such a choice by regarding all interference that is not controlled by explicit synchronization as catastrophic. It is based on three principles: - Operations have duration and can overlap one another during execution. - If two overlapping operations touch the same location, the meaning of the program execution is "wrong''. - If, from a given starting state, execution of a program can give "wrong", then no other possibilities need be considered. We will show a small-step grainless semantics due to Reynolds and a large-step version due to Brookes. Then we will conjecture a more abstract version of the semantics. Brief Bio:   John C. Reynolds received a PhD degree (in theoretical physics) from Harvard University in 1961. He has been Assistant and Associate Physicist at Argonne National Laboratory from 1961 to 1970, Professor of Computer Science at Syracuse University from 1970 to 1986, and Professor of Computer Science at Carnegie Mellon University from 1986 to the present. He has also had temporary appointments at Stanford University, Queen Mary and Westfield College, Edinburgh University, INRIA, Imperial College, Bell Laboratories, and Aarhus University. He is a member of the IFIP Working Group 2.3 on Programming Methodology, a former member of Working Group 2.2 on Formal Language Definition, a member of the advisory board of the journal Higher Order and Symbolic Computation, and a former associate editor of Mathematical Structures in Computer Science, the Journal of the ACM, and the Communications of the ACM. He is best known for his work on functional languages, polymorphic typing, Algol-like languages, and separation logic. In 2003, he received the SIGPLAN Programming Language Achievement Award.
September 23       Wednesday 1:00PM                  4-201 Center for Science and Technology   Dr. Tsuhan Chen	Title: Using Context to Understand Images of People: A Graphical-Model Approach Speaker: Professor Tsuhan Chen, Department of Electrical and Computer Engineering, ECE Director, Cornell University Abstract: When we see other humans, we can quickly make judgments regarding many aspects, including their demographic descriptions and identity if they are familiar to us. We can answer questions related to the activities of, emotional states of, and relationships between people in an image. We draw conclusions based not just on what we see, but also from a lifetime of experience of living and interacting with other people. In this talk, we propose contextual features and graphical models for understanding images of people with the objective of providing computers with access to the same contextual information that humans use. Bio: Professor Chen earned his B.S. from National Taiwan University and his M.S. and Ph.D. from Caltech, all in electrical engineering. After working for Bell Labs for several years, he joined the ECE faculty at Carnegie Mellon University (CMU) in 1997. There, in addition to research and teaching responsibilities, he has served as associate department head of ECE and co-director of the Industrial Technology Research Institute (ITRI) Laboratory at Carnegie Mellon, a collaborative research program with ITRI in Taiwan. In 2007, Chen was elected a fellow of IEEE for contributions in the area of multidisciplinary multimedia signal processing. In 2004 and 2008, he delivered the keynote address at International Conference on Multimedia and Expo (ICME), IEEE’s flagship conference on multimedia technologies. From 2002 to 2004, he served as editor-in-chief of the IEEE Transactions on Multimedia, a publication designed to integrate all aspects of multimedia systems and technology, signal processing, and applications. Chen received the Benjamin Richard Teare Teaching Award in 2006 from the CMU College of Engineering for his consistent excellence in graduate and undergraduate education. He was cited particularly for his success with Signals and Systems, a large undergraduate course serving students with diverse backgrounds and interests, in which he provided clear explanations of complex mathematical methods for analyzing signals and systems. Chen’s success at the intersection of research and education is evident in the awards that he has earned jointly with his graduate students. These include several best paper awards at IEEE Conferences on Computer Vision and Pattern Recognition. Chen was chosen to become the director of the School of Electrical and Computer Engineering (ECE) at Cornell University after an extensive national search. He succeeds Clifford Pollock effective January 2009.
September 30 Wednesday 1:00PM                  4-201 Center for Science and Technology Dr. Hui Xiong	Title: Financial Fraud Detection and Prevention with Data Mining Techniques Speaker: Dr. Hui Xiong, Associate Professor Management Science and Information Systems Department Rutgers, the State University of New Jersey Abstract: Recent years have witnessed increased interests in financial fraud detection and prevention. This is driven by the ever-worsening financial crisis and an increased awareness of the importance of financial risk management. Indeed, the wide availability of fine-grained financial data enables unprecedent opportunities to change the computing paradigm for financial fraud detection and prevention. However, as these financial data become more detailed and multi-dimensional, it becomes ever more difficult for analysts to sift through the data even though it may contain valuable information. Data Mining holds great promise to address this challenge by providing efficient techniques to uncover useful information hidden in the large data repositories. Along this line, in this talk, we focus on introducing the unique features that distinguish data mining techniques from traditional analytic techniques for fraud detection and prevention. Also, as a pilot feasibility study, we will present some real-world case studies to illustrate the applications of data mining techniques for financial fraud detection and prevention. Finally, an examination of major research needs in exploiting data mining techniques for fraud detection and prevention reveals some new opportunities for bio-inspired collaborative fraud detection and prevention in multi-source and multi-level financial data. Bio: Dr. Hui Xiong received his Ph.D. in Computer Science from the University of Minnesota. He is currently an Associate Professor at Rutgers University, where he received a two-year early promotion/tenure (2009),the Rutgers University Board of Trustees Research Fellowship for Scholarly Excellence (2009), an IBM ESA Innovation Award (2008), the Junior Faculty Teaching Excellence Award (2007) and the Junior Faculty Research Award (2008) at the Rutgers Business School. His general area of research is data and knowledge engineering, with a focus on developing effective and efficient data analysis techniques for emerging data intensive applications. He has published prolifically in refereed journals and conference proceedings (3 books, 20+ journal papers, and 40+ conference papers), such as JOC?TKDE?VLDBJ?JDMKD?KDD. He is the co-editor of Clustering and Information Retrieval and the co-Editor-in-Chief of Encyclopedia of GIS. He is an Associate Editor of the Knowledge and Information Systems journal. He has served regularly in the organization committees and the program committees of a number of international conferences and workshops, such as AAAI, KDD, ICDE, CIKM,and ICML. He is a senior member of the IEEE and a member of the ACM.
October 7                                        Wednesday 1:00PM                  4-201 Center for Science and Technology Dr. Subhash Kak	Title: "Prospects for Quantum Computing" Speaker: Dr. Subhash Kak is professor and head of Computer Science Department at Oklahoma State University in Stillwater, OK Abstract: The idea of quantum computing is attractive. It enriches computing theory beyond classical techniques in doing Fourier transform faster (albeit the solution is not fully available), factorization in log n rather than steps, search for an item with a specific property in an unordered database in rather than n/2 steps, solving Pell’s equation in polynomial time, and it has applications to cryptography. But even in theory, quantum computing can only solve problems where the solution amounts a rotation in an appropriate space (as in primality testing since integers form a multiplicative group). The allure of finding new problems that could be shown to be solved faster by quantum algorithms and the challenge of building quantum computing machines has made the field popular amongst mathematicians and physicists. It has also spurred research in materials science and in the experimental area of finding good candidates for implementing quantum circuits. In this paper, we will show that in addition to problems of decoherence and error correction, there exist other fundamental difficulties in implementing quantum gates (and of testing them) that make the quantum circuit model of computing impractical at this time. It is no wonder that progress in physical implementations is limited and there is no unanimity on what kind of physical qubits to use. Researchers have listed five requirements for successful physical implementation and a qubit system that would meet each of these requirements in an easy manner remains to be identified. Even quantum cryptography, which has been physically demonstrated, is not likely to be competitive with classical cryptography. The main advantage of the BB84 protocol over classical cryptography is that eavesdropping can be detected in theory. In practice, lasers that are used for the implementation of BB84 do not transmit single photons and, therefore, BB84 is susceptible to siphoning attack. Although the three-stage quantum cryptography protocol is not susceptible to the siphoning attack, the cost of sending the qubits over three hops is likely to be so high so as to make it unattractive compared to classical systems. The pessimistic assessment of the quantum circuit model of computing does not imply that some other more attractive computing paradigm based on quantum theory does not exist. Bio: Subhash Kak is professor and head of computer science department at Oklahoma State University in Stillwater. He works in the fields of AI, cryptography, quantum information science, and neural networks. He completed his Ph.D. in Electrical Engineering from Indian Institute of Technology, Delhi. During 1979-2007, he was with Louisiana State University, Baton Rouge where he served most recently as Donald C. and Elaine T. Delaune Distinguished Professor of Electrical and Computer Engineering. He is the author of 20 books of which the most recent is "The Architecture of Knowledge." These books inlcude 6 books of verse. His books have been translated into French, German, Italian, Spanish, Korean, and Serbian. Amongst his awards include British Council Fellow (1976), Science Academy Medal of the Indian National Science Academy (1977), Kothari Prize (1977), UNESCO Tokten Award (1986), Goyal Prize (1998), National Fellow of the Indian Institute of Advanced Study (2001), and Distinguished Alumnus of IIT Delhi (2002).
October 14 Wednesday 1:00PM                  4-201 Center for Science and Technology    Dr. Chai Wah Wu	Title: Synchronization and consensus in complex networks of dynamical systems Speaker:  Dr. Chai Wah Wu is currently Manager of the Distributed Digital Media Infrastructure group in the Internet Infrastructure and Computing Utility department at IBM T. J. Watson Research Center. Abstract:  Complex networks of dynamical systems can exhibit order in the form of synchronization amidst apparent chaos. Flocking or swarming behavior in animals and social insects are examples of such behavior. Recently, the problem of coordinating behavior in wireless autonomous robots with localized interaction has also been studied in this context. In this talk, I explore the interplay between algebraic and random graph theory, dynamical systems and linear algebra in analyzing the synchronization and consensus behavior in networks of coupled dynamical systems. In particular, the relationship between graph-theoretical properties and the propensity of the network to synchronize will be discussed. Furthermore, features of complex networks such as small world networks and scale-free networks and how they impact the synchrony of the network will be studied. Finally, applications to consensus and agreement in distributed autonomous agents will be discussed.  Bio:  Chai Wah Wu is Manager of the Distributed Digital Media Infrastructure group at IBM T. J. Watson Research Center in Hawthorne, NY. He received a B.A. degree in Cognitive Science from Lehigh University and a Ph.D. degree in Electrical Engineering from the University of California, Berkeley and has worked on various research topics including image halftoning, multimedia security and synchronization in networks of coupled systems. He has authored over 100 papers and over 50 issued patents. He was elected Fellow of the IEEE in 2001.
October 28 Wednesday 1:00PM                  4-201 Center for Science and Technology Dr. Lizhong Zheng	Title: Dynamic Information Transmission over Feedback Channels Speaker: Dr. Lizhong Zheng, the Department of Electrical Engineering and Computer Sciences and the Laboratory of Information and Decision Systems at Massachusetts Institute of Technology Abstract: We study the classical problem of communication over channels with perfect causal feedbacks. We focus on using the feedback information to improve the reliability of information transmissions. The key issue is to incrementally adjust the coding function according to what the receiver has learned at each time point. The coding problem is formulated as a dynamic programming, with approximated smoothed value functions. While our solutions improve achievable error exponent for general feedback channels, we also raise through this study the fundamental question as how to measure the effectiveness of a dynamic information transmission scheme. Bio: Lizhong Zheng received the B.S and M.S. degrees, in 1994 and 1997 respectively, from the Department of Electronic Engineering, Tsinghua University, China, , and the Ph.D. degree, in 2002, from the Department of Electrical Engineering and Computer Sciences, University of California, Berkeley. Since 2002, he has been working in the Department of Electrical Engineering and Computer Sciences at Massachusetts Institute of Technology, where he is currently the Steven and Rene Finn associate professor. His research interests include information theory, wireless communications and wireless networks. He received Eli Jury award from UC Berkeley in 2002, IEEE Information Theory Society Paper Award in 2003, NSF CAREER award in 2004, AFOSR Young Investigator Award in 2007.
November 18 Wednesday 1:00PM                  4-201 Center for Science and Technology Dr. Elza Erkip	Title:  Cooperative Wireless Networking: From Theory to Practice Speaker:  Professor Elza Erkip, Associate Professor Department of Electrical and Computer Engineering Polytechnic Institute of NYU Abstract: Wireless will be the dominant mode of internet access for end users in near future. However, bandwidth limitations of the wireless channel, interference from multiple users operating in the same band and channel variations due to fading become bottlenecks for typical multimedia applications that require high bandwidth and an error resilient communication medium. Cooperative networking, by enabling wireless terminals to assist each other in transmitting information to their desired destinations, provides a promising technology for improving the performance of wireless networks.          In this talk, we provide an overview or cooperative wireless networking, and summarize some of our recent research activities that span multiple layers of the protocol stack. At the physical layer, we describe how cooperation of users can be used to manage interference, and the interplay between signal relaying and interference forwarding. We discuss design principles of a distributed, robust cooperative medium access control (MAC) layer, and illustrate how benefits of cooperative communications can be realized in a large network. We also outline our current efforts in building a large scale experimental cooperative networking testbed. Bio: Elza Erkip received the the B.S. degree in Electrical and Electronic Engineering from Middle East Technical University, Turkey, and the M.S. and Ph.D. degrees in Electrical Engineering from Stanford University, Stanford, CA. She joined Polytechnic Institute of NYU in Spring 2000, where she is currently an Associate Professor of Electrical and Computer Engineering. She spent 1996-1999 at Rice University, and 2007-2008 academic year at Princeton University. Her general research interests are in wireless communications, information theory and communication theory.           Elza Erkip received the NSF CAREER Award in 2001, the IEEE Communications Society Rice Paper Prize in 2004, and the ICC Communication Theory Symposium Best Paper Award in 2007. She co-authored a paper that received the ISIT Student Paper Award in 2007. Currently, she is an Associate Editor of IEEE Transactions on Information Theory, an Associate Editor of IEEE Transactions on Communications and a Co-Chair of GLOBECOM 2009 Communication Theory Symposium. She was a Publications Editor of IEEE Transactions on Information Theory during 2006-2009, a Guest Editor of IEEE Signal Processing Magazine in 2007, the Publications Chair of ITW 2009, Taormina, the MIMO Communications and Signal Processing Technical Area Chair of Asilomar Conference on Signals, Systems, and Computers in 2007, and a Technical Program Co-Chair of Communication Theory Workshop in 2006. _____________________________ Jointly Sponsored by Department of Electrical Engineering and Computer Science and CASE Center http://www.lcs.syr.edu/academic/dept_electricalengcompsci/colloquium.aspx

Colloquium Archive