Abdellatif ZAIDI - Communication systems
Date: - Location: Eurecom
Title: Statistical Learning via Information Bottleneck Abstract: We connect the information flow in a neural network to sufficient statistics; and show how techniques that are rooted in information theory, such as the source-coding based information bottleneck method can lead to improved architectures, as well as a better understanding of the theoretical foundation of neural networks, viewed as a cascade compression networks. We illustrate our results and view through some numerical examples. Biography: A. Zaidi received the B.S. degree in Electrical Engineering from ENSTA ParisTech, Paris, in 2002 and the M. Sc. and Ph.D. degrees in Electrical Engineering from TELECOM ParisTech, Paris, France in 2002 and 2006, respectively. From December 2002 to March 2006, he was with the Communications an Electronics Dept., TELECOM ParisTech, Paris, France and the Signals and Systems Lab., CNRS/Supélec, France pursuing his PhD degree. From May 2006 to September 2010, he was at École Polytechnique de Louvain, Université Catholique de Louvain, Belgium, working as a senior researcher. Dr. Zaidi was "Research Visitor" at the University of Notre Dame, Indiana, USA, during 2007 and 2008, the Technical University of Munich during Summer 2014, and the Ecole Polytechnique Federale de Lausanne, EFPL, Switzerland. He is Associate Professor at University Paris-Est, France; and on leave since Jan. 2015 at the Mathematics and Algorithmic Sciences Lab., France Research Center, Huawei France. His research interests lie broadly in network information theory and its interactions with other fields, including communication and coding, statistics, security and privacy, with application to diverse problems of data transmission and compression in networks. Dr. Zaidi is IEEE senior member. From 2013 to 2016 he served as Associate Editor for the Eurasip Journal on Wireless Communications and Networking (EURASIP JWCN); and, since 2016, as Associate Editor for the IEEE Transactions on Wireless Communications. He is the co-recipient (jointly with Shlomo Shamai (Shitz)) of the N# Best Paper Award, as well as the French Excellence in Research Award since 2011.
