{"id":11252,"date":"2024-10-14T13:22:28","date_gmt":"2024-10-14T11:22:28","guid":{"rendered":"https:\/\/www.isir.upmc.fr\/?post_type=isir_news&#038;p=11252"},"modified":"2025-02-07T10:37:25","modified_gmt":"2025-02-07T09:37:25","slug":"ingredients-for-motion-planning-powered-reinforcement-learning-soutenance-hdr-de-nicolas-perrin-gilbert","status":"publish","type":"isir_news","link":"https:\/\/www.isir.upmc.fr\/actualites\/ingredients-for-motion-planning-powered-reinforcement-learning-soutenance-hdr-de-nicolas-perrin-gilbert\/","title":{"rendered":"\u00ab\u00a0Ingredients for Motion Planning-powered Reinforcement Learning\u00a0\u00bb, soutenance HDR de Nicolas Perrin-Gilbert"},"content":{"rendered":"\n<div class=\"wp-block-media-text is-stacked-on-mobile\" style=\"grid-template-columns:26% auto\"><figure class=\"wp-block-media-text__media\"><img loading=\"lazy\" decoding=\"async\" width=\"391\" height=\"391\" src=\"https:\/\/www.isir.upmc.fr\/wp-content\/uploads\/2022\/06\/photo_perso_visage.jpg\" alt=\"\" class=\"wp-image-8257 size-full\" srcset=\"https:\/\/www.isir.upmc.fr\/wp-content\/uploads\/2022\/06\/photo_perso_visage.jpg 391w, https:\/\/www.isir.upmc.fr\/wp-content\/uploads\/2022\/06\/photo_perso_visage-300x300.jpg 300w, https:\/\/www.isir.upmc.fr\/wp-content\/uploads\/2022\/06\/photo_perso_visage-150x150.jpg 150w\" sizes=\"auto, (max-width: 391px) 100vw, 391px\" \/><\/figure><div class=\"wp-block-media-text__content\">\n<p class=\"wp-block-paragraph\"><strong>Nicolas Perrin-Gilbert<\/strong>, charg\u00e9 de recherche CNRS, soutiendra son Habilitation \u00e0 Diriger des Recherches (HDR), le jeudi 17 octobre \u00e0 16h30 \u00e0 l&rsquo;ISIR, sur le campus Pierre et Marie Curie de Sorbonne Universit\u00e9.&nbsp;<\/p>\n<\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">Titre de ses travaux :&nbsp;<strong>\u201cIngredients for Motion Planning-powered Reinforcement Learning\u201d<\/strong>.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">La constitution du jury est la suivante&nbsp;:&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">&#8211; Aleksandra Faust, Google Deepmind (Rapportrice),<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">&#8211; Matthieu Geist, Universit\u00e9 de Lorraine &amp; Cohere (Examinateur),<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">&#8211; Nicolas Mansard, LAAS-CNRS (Rapporteur),<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">&#8211; Jochen J. Steil, Universit\u00e9 Technique de Brunswick (Rapporteur),<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">&#8211; Nicolas Thome, Sorbonne Universit\u00e9 (Examinateur)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>R\u00e9sum\u00e9 de son habilitation&nbsp;:&nbsp;<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Mon HDR se concentre sur quatre contributions principales, assez distinctes, mais ayant toutes un lien avec un objectif commun, celui d&rsquo;am\u00e9liorer l&rsquo;exploration en apprentissage par renforcement \u00e0 travers l&rsquo;utilisation de techniques de planification de mouvements bas\u00e9es sur l&rsquo;\u00e9chantillonnage al\u00e9atoire.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">La premi\u00e8re contribution concerne des transformations g\u00e9om\u00e9triques entre mouvements continus et s\u00e9quences discr\u00e8tes de contacts, ce qui permet d&rsquo;aborder les probl\u00e8mes de g\u00e9n\u00e9ration de mouvements de locomotion sous un angle nouveau. La deuxi\u00e8me contribution propose une technique de g\u00e9n\u00e9ralisation ou d&rsquo;imitation de trajectoires bas\u00e9e sur l&rsquo;application de transformations diff\u00e9omorphiques successives. La troisi\u00e8me porte sur la gestion de s\u00e9quences successives d&rsquo;objectifs, en particulier sur les notions de buts interm\u00e9diaires et les fa\u00e7ons de les consid\u00e9rer pour atteindre plus efficacement un objectif global. Enfin, la quatri\u00e8me contribution pr\u00e9sente un algorithme d&rsquo;apprentissage par renforcement dit \u00ab\u00a0off-policy\u00a0\u00bb, visant \u00e0 am\u00e9liorer l&rsquo;entra\u00eenement de politiques de contr\u00f4le lorsqu&rsquo;une partie significatives des donn\u00e9es provient de trajectoires d&rsquo;exploration.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Contact&nbsp;:&nbsp;<\/strong><a href=\"https:\/\/www.isir.upmc.fr\/personnel\/perrin\/\" target=\"_blank\" rel=\"noreferrer noopener\">Nicolas Perrin-Gilbert<\/a>, charg\u00e9 de recherche CNRS<\/p>\n\n\n\n<hr class=\"wp-block-separator has-text-color has-white-color has-alpha-channel-opacity has-white-background-color has-background\"\/>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Publi\u00e9 le 14\/10\/2024. <\/em><\/p>\n","protected":false},"featured_media":10014,"template":"","isir_news_categories":[260],"class_list":["post-11252","isir_news","type-isir_news","status-publish","has-post-thumbnail","hentry","isir_news_categories-soutenance"],"_links":{"self":[{"href":"https:\/\/www.isir.upmc.fr\/wp-json\/wp\/v2\/isir_news\/11252","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.isir.upmc.fr\/wp-json\/wp\/v2\/isir_news"}],"about":[{"href":"https:\/\/www.isir.upmc.fr\/wp-json\/wp\/v2\/types\/isir_news"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.isir.upmc.fr\/wp-json\/wp\/v2\/media\/10014"}],"wp:attachment":[{"href":"https:\/\/www.isir.upmc.fr\/wp-json\/wp\/v2\/media?parent=11252"}],"wp:term":[{"taxonomy":"isir_news_categories","embeddable":true,"href":"https:\/\/www.isir.upmc.fr\/wp-json\/wp\/v2\/isir_news_categories?post=11252"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}