The dual geometry of Shannon information and its applications

4,082 views · Published 7 November 2016 · 1:04:56 · Indexed 14 November 2026

Channel: Institut Henri Poincaré · 2016 · Science & Technology

Watch on YouTube

Théorie de l'information : nouvelles frontières dans le cadre du Centenaire de Claude Shannon 
By Frank Nielsen (École Polytechnique)
Abstract: In information geometry, the negative Shannon entropy, called the Shannon information, is a strictly convex and differentiable function that induces a dually flat manifold structure equipped with the Kullback- Leibler divergence. In this talk, I review the concept of dual geometries, introduce the dual space of spheres, and describe the role of divergences in information theory, statistics, pattern recognition and machine learning.

Frank Nielsen received his PhD (1996) and his habilitation (2006) on computational geometry from the University of Nice-Sophia Antipolis, France. After the French national service, he joined Sony CSL (Japan) in 1997. He is currently professor in the computer science department of École Polytechnique (France). He co- organizes with Frédéric Barbaresco (Thales) the biannual Geometric Sciences of Information (GSI) conference, and is an associate editor of the Springer Journal of Mathematical Imaging and Vision and of MDPI Entropy.

More from this channel