Data Institute Seminar: Introduction to graph signal processing by Pierre Borgnat from ENS Lyon

on the October 5, 2017

Thursday 5th of October at 14:30
Pierre Borgnat from ENS Lyon will give a talk on “An invitation to Graph Signal Processing with Applications to Clustering”. The event will take place on Thursday 5 October at GIPSA-lab.
In the recent years, methods and topics from signal (and image) processing were developed so as to address problems in data science on networks of graphs. Among works in this domain, seminal results were the developement of Fourier transforms for graph signals, of multiscale decompositions, of graph filterbanks, of sampling results on graphs,... In this talk, we will present a broad introduction to this domain and the core elements and concepts of graph signal processing. We will also discuss works and developments of graph signal processing, especially those aiming at revising graph-based clustering.

Pierre Borgnat is a CNRS Directeur de Recherche, at the Laboratory of Physics, ENS de Lyon. Born in France in 1974, he was received at the École Normale Supérieure de Lyon in 1994 and as a Professeur Agrégé in Physical Sciences in 1997. He received the Ph.D. degree in Physics and Signal Processing in 2002. He is in CNRS since 2004 (and DR since 2016) and he is currently director of IXXI (Complex System Institute of Rhône-Alpes). His research interests are in statistical signal processing, mainly graph signal processing, complex networks, nonstationary signals or scaling phenomena. He works on several applications of these signal processing methods: Internet traffic modeling and measurements, fluid mechanics, analysis of social data, transportation studies.

Seminar organised by Sophie Achard, GIPSA-lab

Published on October 3, 2017

Practical informations


GIPSA-lab, bâtiment D, salle Mont-Blanc