This school « Statistical and Geometric divergences for Machine Learning, part 2 » will take place at Rennes in Villejean campus rom June 20 to June 24, 2022. This school explores mathematical foundations of machine learning and deep learning with a special focus on entropies, divergences and distances used in machine learning. The school is intended for Master's degree and PhD students and young researchers as well as senior researchers interested in these topics.
It will offer four mini-lectures given by confirmed researchers in the most recent fields of the mathematics of Artificial Intelligence:
- Lecture 1 : Kernel based distances and applications in statistics and machine learning (Bharath Sriperumbudur, Pennsylvania State University)
- Lecture 2 : Kernel and optimal transport-based Generative Adversarial Neural networks : Arthur Gretton (Univ. College London, Gatsby Computational Neuroscience Unit)
- Lecture 3 : Optimal transport distances and domain adaptation : Laetitia Chapel (Univ. Bretagne Sud), Nicolas Courty (Univ. Bretagne Sud), Rémi Flamary (CMAP, École Polytechnique)
- Lecture 4 : Optimal transport and fair learning : Jean-Michel Loubes (Univ. Toulouse Paul Sabatier)
and complementary presentations.
Key words: Reproducing Kernel Hilbert Spaces, kernel methods, energy distances, maximum mean discrepancy, optimal transport, generative adversarial networks.