BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Information-optimal drifts and the geometry of Markov processes
DTSTART:20151203T171500
DTEND:20151203T181500
DTSTAMP:20260427T225732Z
UID:b923b13c8aba5614fe0b0e889b30e129bf613a84084d3b5f3d701155
CATEGORIES:Conferences - Seminars
DESCRIPTION:James R. Lee (University of Washington)\nIf one records a Mark
 ov process converging to stationarity from an initial distribution and the
 n plays the video in reverse\, what appears is a new (time inhomogeneous) 
 Markov process that constructs the initial distribution using as little in
 formation as possible. One can leverage this entropic optimality to study 
 the behavior of the random walk\, especially at "early" times.  For some 
 problems\, this perspective appears to offer something new over semi-group
  methods.  We discuss applications to regularizing properties of classica
 l diffusion operators and to functional inequalities on finite graphs and 
 groups.  There are interesting connections to discrete analogs of no
LOCATION:BI A0 448 https://plan.epfl.ch/?room==BI%20A0%20448
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR
