BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Information theory with kernel methods
DTSTART:20240207T131500
DTEND:20240207T151500
DTSTAMP:20260407T182225Z
UID:f20c79232d1facd80cabafd4e6b2048d459e1a2e598afa09cbaad11a
CATEGORIES:Conferences - Seminars
DESCRIPTION:Francis Bach    \nEstimating and computing entropies of pr
 obability distributions are key computational tasks throughout data scienc
 e. In many situations the underlying distributions are only known through 
 the expectation of some feature vectors which has led to a series of works
  within kernel methods with applications to generative modeling and probab
 ilistic inference. In this talk I will explore the particular situation wh
 ere the feature vector is a rank-one positive definite matrix and show how
  the associated expectations (a covariance matrix) can be used with inform
 ation divergences from quantum information theory to draw direct links wit
 h the classical notions of Shannon entropies.
LOCATION:CM 1 4 https://plan.epfl.ch/?room==CM%201%204
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR