{"id":394,"date":"2019-04-14T16:28:20","date_gmt":"2019-04-14T15:28:20","guid":{"rendered":"http:\/\/pageperso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/?page_id=394"},"modified":"2019-06-27T14:12:21","modified_gmt":"2019-06-27T13:12:21","slug":"third-week","status":"publish","type":"page","link":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/accueil\/master-school-on-data-science-and-geometry\/third-week\/","title":{"rendered":"Third Week 15-19 July"},"content":{"rendered":"<p><span style=\"color: #ff99cc;font-size: 14pt\">Geometry Lecture (10h) <em>Wasserstein geometry and optimal transport<\/em><\/span><\/p>\n<p><span style=\"font-size: 14pt\">Max Fathi<\/span><\/p>\n<ul>\n<li><span style=\"font-size: 14pt\">Lecture 1 (2 hours) Introduction to the optimal transport problem on Euclidean space. Formulations of Monge and Kantorovitch, history, applications. Explicit solution in dimension one. Existence of solutions to the Kantorovitch problem.<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 2(2\u00a0hours)\u00a0 Kantorovitch duality, existence of a transport map solving the Monge problem. Connection with the Monge-Ampere PDE. Extension to Riemannian manifolds.<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 3 (2\u00a0hours)\u00a0Transport cost as a distance on the space of probability measures, and applications in statistics.<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 4 (2\u00a0hours)\u00a0 The geometry of optimal transport: Benamou-Brenier formula and Riemannian structure of the space of probability measures. Application: gradient flow structure of the heat equation.<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 5 (2\u00a0hours)\u00a0 Long-time behavior of stochastic processes, and applications to numerical schemes.<\/span><\/li>\n<\/ul>\n<p><span style=\"color: #ff99cc;font-size: 14pt\">Statistics Lecture (10h) <em>Using the geometry of the Wasserstein space\u00a0 in statistical learning<\/em><\/span><\/p>\n<p><span style=\"font-size: 14pt\">Alice Le Brigant<\/span><\/p>\n<ul>\n<li><span style=\"font-size: 14pt\">Lecture 1 (2\u00a0 hours)\u00a0 The Fisher information in statistical inference;\u00a0 Kullback-Leibler divergence; Search for the best estimator: Cram\u00e9r-Rao\u00a0 bound and efficient estimators.<br \/>\n<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 2 (2 hours) The Fisher information metric; Fisher geometry of Gaussian distributions; Fr\u00e9chet mean and the Karcher flow algorithm.<br \/>\n<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 3 (2 hours) Information geometry manifold; Dual affine connections; Information alpha manifolds; Divergences; Geometric EM algorithm.<br \/>\n<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 4 (2 hours) PCA for a closed convex subset of a Hilbert space; The Wasserstein space in dimension 1, the exponential and\u00a0 logarithm maps, PCA for a closed convex subset of\u00a0 an Hilbert Spaces, functional PCA in the Wasserstein Space.<br \/>\n<\/span><\/li>\n<li><span style=\"font-size: 14pt\">Lecture 5 (2 hours) Wasserstein Barycentre; Existence and uniqueness\u00a0 of empirical and population Wasserstein barycentre; Minimax convergence\u00a0 rate for estimating the Wasserstein barycentre of random measures on the real line.<br \/>\n<\/span><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Geometry Lecture (10h) Wasserstein geometry and optimal transport Max Fathi Lecture 1 (2 hours) Introduction to the optimal transport problem on Euclidean space. Formulations of Monge and Kantorovitch, history, applications. Explicit solution in dimension one. Existence of solutions to the Kantorovitch problem. Lecture 2(2\u00a0hours)\u00a0 Kantorovitch duality, existence of a transport map solving the Monge problem. [&hellip;]<\/p>\n","protected":false},"author":203,"featured_media":0,"parent":377,"menu_order":0,"comment_status":"closed","ping_status":"open","template":"template-full-width.php","meta":{"footnotes":""},"class_list":["post-394","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/pages\/394","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/users\/203"}],"replies":[{"embeddable":true,"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/comments?post=394"}],"version-history":[{"count":6,"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/pages\/394\/revisions"}],"predecessor-version":[{"id":404,"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/pages\/394\/revisions\/404"}],"up":[{"embeddable":true,"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/pages\/377"}],"wp:attachment":[{"href":"https:\/\/perso.math.univ-toulouse.fr\/statistics-geometry-and-topology\/wp-json\/wp\/v2\/media?parent=394"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}