Graduate Student / Postdoc Seminar

Disequilibrium, Coarse-Graining and Information

Speaker: Richard Kleeman

Location: Warren Weaver Hall 1302

Date: Friday, November 15, 2013, 1 p.m.


Dynamical systems involving a large number of degrees of freedom are a practical reality as well as a large challenge mathematically. A productive appoach is to regard them statistically and to consider only gross variables which vary relatively slowly. As time proceeds forward such systems often equilibrate in that mean quantities reach steady values. This process has been (and stiil is) a subject with rather profound implications for the idea of time. In this talk I look at work I have done in this area over the past few years. My approach has been rooted in the field of information theory which provides an intuitive and fundamental set of functionals for viewing the statistical nature of such systems. The nature of predictability of these systems can be analyzed productively this way and related to the idea of the degree of system disequilibrium. The nature of the equilibration process is an area without a universal framework and many approaches have been attempted in the past 50 years. Such a situation contrasts markedly with that concerning systems in equilibrium. I will describe some recent work by myself and a co-worker (Turkington) which takes a novel approach to the challenging problem of system out of equilibrium.