BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Prof. Stéphane Mallat: Modeling Deep Networks: Network Learning f
 or Image Processing
DTSTART:20211118T170000
DTEND:20211118T180000
DTSTAMP:20260413T111136Z
UID:d7b38ddcd523409f0c4481dd61455e3456838b32b7bacdaa24ae8d24
CATEGORIES:Conferences - Seminars
DESCRIPTION:Stéphane Mallat\, Collège de France\nThis event is part of t
 he EPFL Seminar Series in Imaging\n\nAbstract. Deep neural network learni
 ng from data has taken over image processing. Not just for classificaiton 
 and regression but also for denoising and inverse problems. Is it the end 
 of geometric models and understanding ? Deep network models are high-dimen
 sional and must be analyzed in a probabilistic framework. Yet they must al
 so take into account image properties\, including multiscale structures an
 d symmetries. The lecture takes an information theory point of view\, and 
 shows that the underlying mathematics are closely related to statistical p
 hysics. We introduce a general class of interpretable neural network model
 s through the renormalisation group and multiscale wavelet transforms\, wi
 th applications to image generation and classification.\n\nBiography. Sté
 phane Mallat’s research interests include machine learning\, signal proc
 essing\, and harmonic analysis. Starting from truly original theoretical w
 ork\, he has developed their applications up to industrial transfer\, with
  10 international patents. He holds a Ph.D. from the University of Pennsyl
 vania. Since 2017\, he has held the “Data Sciences” chair at the Coll
 ège de France. From 2001 to 2007 he was co-founder and CEO of a semicondu
 ctor start-up company which has grown into a semiconductor company manufac
 turing millions of electronic chips to increase the resolution of pictures
  in high definition televisions. He is a member of the French Academy of s
 ciences\, a foreign member of the US National Academy of Engineering\, an 
 IEEE Fellow and an EUSIPCO Fellow.
LOCATION:https://epfl.zoom.us/j/65087063882
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR