Tuesday, 19th of February

14h30 (room R2014, 660 building) (see location)

Loris Felardos

(TAU team / IBPC)

Title: An Introduction to Graph Neural Networks

Abstract

A considerable amount of learning tasks rely on data that can
be expressed in the form of graphs (including molecules, interactomes,
syntax trees, knowledge graphs, neural networks and brain connectomes).
Several difficulties arise when trying to apply deep learning methods to
graph data (like edge invariance, over-smoothing, or the lack of clear
pooling functions) and a significant amount of research efforts has been
devoted to this area. In this talk, I'll give an introduction to graph
neural networks, their main challenges as well as several methods that
have been proposed to address them.



Contact: guillaume.charpiat at inria.fr
All TAU seminars: here