{"id":2875548,"date":"2015-11-29T09:33:12","date_gmt":"2015-11-29T15:33:12","guid":{"rendered":"http:\/\/hunch.net\/?p=2875548"},"modified":"2015-12-09T12:33:39","modified_gmt":"2015-12-09T18:33:39","slug":"cntk-and-vowpal-wabbit-tutorials-at-nips","status":"publish","type":"post","link":"https:\/\/hunch.net\/?p=2875548","title":{"rendered":"CNTK and Vowpal Wabbit tutorials at NIPS"},"content":{"rendered":"<p>Both <a href=\"https:\/\/cntk.codeplex.com\/documentation\">CNTK<\/a> and <a href=\"https:\/\/hunch.net\/~vw\">Vowpal Wabbit<\/a> have pirate tutorials at <a href=\"http:\/\/nips.cc\">NIPS<\/a>.  The CNTK tutorial is 1 hour during the lunch break of the <a href=\"http:\/\/opt-ml.org\/schedule.html\">Optimization workshop<\/a> while the VW tutorial is 1 hour during the lunch break of the <a href=\"http:\/\/research.microsoft.com\/en-us\/um\/people\/manik\/events\/xc15\/schedule.html\">Extreme Multiclass<\/a> workshop.  Consider dropping by either if interested.<\/p>\n<p>CNTK is a deep learning system started by the speech people who <a href=\"http:\/\/citeseerx.ist.psu.edu\/viewdoc\/download?doi=10.1.1.185.1908&#038;rep=rep1&#038;type=pdf\">started the deep learning craze<\/a> and grown into a more general platform-independent deep learning system.  It has various useful features, the most interesting of which is perhaps efficient scalable training.  Using GPUs with allreduce and <a href=\"http:\/\/research.microsoft.com\/apps\/pubs\/default.aspx?id=230137\">one-bit sgd<\/a> it achieves both high efficiency and scalability over many more GPUs than could ever fit into a single machine.  This capability is unique amongst all open deep learning codebases so everything else looks nerfed in comparison.  CNTK was <a href=\"http:\/\/cntk.codeplex.com\/\">released in April<\/a> so this is the first chance for many people to learn about it.  See <a href=\"blogs.technet.com\/b\/inside_microsoft_research\/archive\/2015\/12\/07\/microsoft-computational-network-toolkit-offers-most-efficient-distributed-deep-learning-computational-performance.aspx\">here<\/a> for more details.<\/p>\n<p>The Vowpal Wabbit tutorial just focuses on what is new this year.  <\/p>\n<ol>\n<li>The learning to search framework has greatly matured and is now easily used to solve ad-hoc joint(structured) prediction problems.  The <a href=\"https:\/\/hunch.net\/~l2s\">ICML tutorial<\/a> covers algorithms\/analysis so this is about using the system.<\/li>\n<li>  VW has also become the modeling element of a <a href=\"http:\/\/research.microsoft.com\/en-us\/projects\/mwt\/\">larger <em>system<\/em><\/a> (called the decision service) which gathers data and uses it as per  <a href=\"https:\/\/hunch.net\/~jl\/interact.pdf\">Contextual Bandit<\/a> learning.  This is now generally usable, and is the first general purpose system of this sort.<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>Both CNTK and Vowpal Wabbit have pirate tutorials at NIPS. The CNTK tutorial is 1 hour during the lunch break of the Optimization workshop while the VW tutorial is 1 hour during the lunch break of the Extreme Multiclass workshop. Consider dropping by either if interested. CNTK is a deep learning system started by the &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/hunch.net\/?p=2875548\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;CNTK and Vowpal Wabbit tutorials at NIPS&#8221;<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[4,42,29],"tags":[],"class_list":["post-2875548","post","type-post","status-publish","format-standard","hentry","category-announcements","category-code","category-machine-learning"],"_links":{"self":[{"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/2875548","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2875548"}],"version-history":[{"count":0,"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/2875548\/revisions"}],"wp:attachment":[{"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2875548"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2875548"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2875548"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}