BIOGRAPHY
I split my research time roughly 50-50 between probability theory and information theory. The more theoretical part of my work is probability. I have been trying to understand the fundamental properties of some of the most important probabilistic models — Markov processes. Although Markov processes have been around for about 100 years and have been used in a huge number of applications, from epidemiology to linguistics and from signal processing to thermodynamics, many basic questions remain unanswered.
In information theory, my main interest is in (lossy) data compression. A solid theoretical basis for data compression exists and it tells us precisely how much we can compress any given data set. There is also an enormous amount of work on developing practical algorithms for data compression. But, so far, there is very little, if anything, connecting the two.
Before coming to Brown, I was an assistant professor of statistics at Purdue University. I did my graduate work in electrical engineering at Stanford. During that time, I spent six months at IBM research, working on a NASA-funded project on satellite image processing and compression.
|
