{"id":138,"date":"2020-10-21T09:15:10","date_gmt":"2020-10-21T07:15:10","guid":{"rendered":"https:\/\/project.inria.fr\/hpcbigdata\/?p=138"},"modified":"2020-10-21T09:15:17","modified_gmt":"2020-10-21T07:15:17","slug":"inria-skoltech-moliere-associated-team","status":"publish","type":"post","link":"https:\/\/project.inria.fr\/hpcbigdata\/inria-skoltech-moliere-associated-team\/","title":{"rendered":"Inria-Skoltech Moliere Associated Team"},"content":{"rendered":"<p><a href=\"https:\/\/team.inria.fr\/moliere\/\">Moli\u00e8re (Memory Optimization for paraLlel traIning of dEep neuRal nEtworks)<\/a> is an Inria Associated Team\u00a0between Hiepacs project team\u00a0\u00a0and the Center for Computational and Data-Intensive Science and Engineering at Skoltech<\/p>\n<p class=\"p1\">The main goal \u00a0concerns the development of algorithms to better manage memory constraints and increase the scalability of DNN training algorithms, by combining Checkpointing and Tensor Train Decomposition techniques with data parallelism and model parallelism.<\/p>\n<h2>Research directions<\/h2>\n<ul>\n<li>Re-materialization (Checkpointing)<\/li>\n<li>Tensor Train (TT) Decomposition<\/li>\n<li>MultiGrid Reduction In Time (MGRIT)<\/li>\n<li>Pipelined Model Parallelism<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Moli\u00e8re (Memory Optimization for paraLlel traIning of dEep neuRal nEtworks) is an Inria Associated Team\u00a0between Hiepacs project team\u00a0\u00a0and the Center for Computational and Data-Intensive Science and Engineering at Skoltech The main goal \u00a0concerns the development of algorithms to better manage memory constraints and increase the scalability of DNN training algorithms,\u2026<\/p>\n<p> <a class=\"continue-reading-link\" href=\"https:\/\/project.inria.fr\/hpcbigdata\/inria-skoltech-moliere-associated-team\/\"><span>Continue reading<\/span><i class=\"crycon-right-dir\"><\/i><\/a> <\/p>\n","protected":false},"author":1416,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","_members_access_role":[],"_members_access_error":""},"categories":[1],"tags":[],"class_list":["post-138","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/posts\/138","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/users\/1416"}],"replies":[{"embeddable":true,"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/comments?post=138"}],"version-history":[{"count":1,"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/posts\/138\/revisions"}],"predecessor-version":[{"id":139,"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/posts\/138\/revisions\/139"}],"wp:attachment":[{"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/media?parent=138"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/categories?post=138"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/project.inria.fr\/hpcbigdata\/wp-json\/wp\/v2\/tags?post=138"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}