BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Joint seminar IPHYS & IEL: Signal processing and Machine Learning 
 in high-dimensions: Computation\, Physics and Optics
DTSTART:20200219T111500
DTEND:20200219T121500
DTSTAMP:20260510T164801Z
UID:f1f0a0d6f853ec2b736d31edf5da8fcea5c796510c539f917d68e5a3
CATEGORIES:Conferences - Seminars
DESCRIPTION:Florent Krzakala is a professor at Sorbonne Université and a 
 Researcher at Ecole Normale Superieure in Paris. His research interests in
 clude Statistical Physics\, Machine Learning\, Statistics\, Signal Process
 ing\, Computer Science and Computational Optics. He leads the SPHINX “St
 atistical PHysics of INformation eXtraction” team in Ecole Normale in Pa
 ris\, and is the holder of the CFM-ENS Datascience chair and of a PRAIR
 IE Institute chair. He is also the funder and scientific advisor of the st
 artup Lighton.  \nAbstract: The increasing dimensionality of data in th
 e modern computing age presents new challenges and opportunities to the fi
 eld of signal and data processing.  One the one hand\, processing these d
 ata requires new software and hardware solutions\, on the other hand\, the
  high-dimensional settings allow one to use powerful asymptotic methods fr
 om probability theory and statistical physics to obtain precise characteri
 zations and develop new algorithmic approaches.\n\nHere I will present som
 e of the results obtained through these methods for two sets of problems: 
 (i) reconstructing a signal from a minimal number of measurements in a var
 iety of situations\, including for instance phase retrieval and (ii) analy
 zing the performances of neural networks and their ability to avoid overfi
 tting. I will also discuss some practical implementations of these ideas f
 or both problems in the context of optics in random media and optical comp
 uting.\n 
LOCATION:BM 5202 https://plan.epfl.ch/?room==BM%205202
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR