CSL Communications Group Calendar
http://illinois.edu/calendar/list/3123
CSL Communications Group CalendarCOMM Seminar - “The Entropy Region and the Four-Atom Conjecture”
http://illinois.edu/calendar/detail/3123/31695135
http://illinois.edu/calendar/detail/3123/31695135Mon, 21 Apr 2014 16:00:00 CDT**Abstract**
The entropy region is central to computing network coding capacities. However, the entropy region for n random variables remains mysterious for n > 3. When n is 4, several outer bounds were found, but they are not tight. The Ingleton scores was introduced to measure the entropy region. This has an upper bound, but its supremum is still unknown. The Four-Atom Conjecture, which said that s < 0.089373, was disproved by Matus and Csirmaz last year. In this talk, we introduce a systematic approach that obtains larger values of s than Matus and Csirmaz, and investigate the true value of its supremum.
****Bio****
Professor Nigel Boston grew up in England and attended Cambridge and Harvard. His postdoctoral work in Paris and Berkeley was followed by 12 years at the University of Illinois, except for six months as Rosenbaum Fellow at the Newton Institute in Cambridge, UK. In recent years he moved towards engineering, becoming founding director of the Illinois Center for Cryptography and Information Protection in the Coordinated Science Laboratory. In 2002, he was hired by the University of Wisconsin - Madison as part of the computational sciences cluster, with joint appointments in Mathematics and Electrical and Computer Engineering. He spent 2006-7 as Hedberg Chair at the University of South Carolina and 2008-9 as Stokes Professor of Pure and Applied Algebra at University College Dublin, Ireland.SPOTLIGHT - COMM Seminar - “Optimal Rate Communication by Regression ”
http://illinois.edu/calendar/detail/3123/31697282
http://illinois.edu/calendar/detail/3123/31697282Mon, 28 Apr 2014 16:00:00 CDT***Abstract***
We discuss our recently developed sparse superposition codes for the Gaussian noise channel. With a fast adaptive successive decoder it achieves nearly exponentially small error probability at any fixed rate R less than the Shannon capacity. This is joint work with Antony Joseph and Sanghee Cho.
******Bio******
Professor Barron's research interests include the areas of statistical information theory, statistical inference, model selection, probability limit theorems, asymptotics of Bayes procedures, curve and surface estimation, artificial neural networks, approximation theory, and investment theory. Received Ph.D., Electrical Engineering, Stanford University; M.S., Electrical Engineering, Stanford University; B.S. (Magna Cum Laude); E.E. and Math Science, Rice University. 1985 – 1992 Andrew was Assistant /Associate Professor of Statistics and Electrical & Computer Engineering, University of Illinois.COMM Seminar - “TBA”
http://illinois.edu/calendar/detail/3123/31697284
http://illinois.edu/calendar/detail/3123/31697284Mon, 05 May 2014 16:00:00 CDT