BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : Scattering Representations for Recogntion
DTSTART:20140306T161500
DTEND:20140306T173000
DTSTAMP:20260508T194300Z
UID:3dee71162d8fbed4f78203ac8351bd264ce9d17bc98f14896a2d8f69
CATEGORIES:Conferences - Seminars
DESCRIPTION:By : Joan Bruna\, New York University\nIC Faculty candidateAbs
 tract\nObject and Texture Classification are fundamental problems in which
  one is required to extract stable\, discriminative information out of noi
 sy\, high-dimensional signals. Our perception of image and audio patterns 
 is invariant under several transformations\, such as illumination changes\
 , translations or frequency transpositions\, as well as small geometrical 
 perturbations. Similarly\, textures are examples of stationary\, non-gauss
 ian\, intermittent processes which can be recognized from few realizations
 . Scattering operators construct a non-linear signal representation by cas
 cading wavelet modulus decompositions\, shown to be\nstable to geometric d
 eformations\, and capturing high-order moments with low-variance estimator
 s. Moreover\, scattering coefficients encode the presence\nof geometric re
 gularity\, modulation phenomena\, intermittency and self-similarity\, lead
 ing to efficient classification\, detection and characterization of severa
 l pattern and multifractal texture recognition tasks. Although stability t
 o geometric perturbations is necessary\, it is not sufficient for the most
  challenging object recognition tasks\, which require learning the invaria
 nce from data. We shall see that scattering operators can be generalized t
 o this scenario\, highlighting the close links between structured dictiona
 ry learning approaches and deep neural networksBiography\nJoan Bruna gradu
 ated from Universitat Politècnica de Catalunya in both Mathematics and El
 ectrical Engineering\, in 2002 and 2004 respectively. He obtained an MSc i
 n applied mathematics from ENS Cachan in 2005. From 2005 to 2012\, he was 
 a research engineer in an image processing startup\, developing realtime\n
 video processing algorithms. In 2013 he obtained his PhD in Applied Mathem
 atics at École Polytechnique\, under the supervision of Prof. Stéphane M
 allat.\nSince fall 2012 he is a postdoctoral researcher in Yann LeCun's la
 b at the Courant Institute\, New York. His research interests include inva
 riant signal representations\, stochastic processes\, harmonic analysis\, 
 deep learning\, and its applications to computer vision.\nMore information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR