{"id":222592,"date":"2014-12-06T20:46:47","date_gmt":"2014-12-07T02:46:47","guid":{"rendered":"http:\/\/hunch.net\/?p=222592"},"modified":"2014-12-06T20:46:47","modified_gmt":"2014-12-07T02:46:47","slug":"vowpal-wabbit-7-8-at-nips","status":"publish","type":"post","link":"https:\/\/hunch.net\/?p=222592","title":{"rendered":"Vowpal Wabbit 7.8 at NIPS"},"content":{"rendered":"<p>I just created <a href=\"https:\/\/github.com\/JohnLangford\/vowpal_wabbit\/tags\">Vowpal Wabbit 7.8<\/a>, and we are planning to have an increasingly less heretical followup tutorial during the non-&#8220;ski break&#8221; at the <a href=\"http:\/\/nips.cc\">NIPS<\/a> <a href=\"http:\/\/opt-ml.org\/schedule.html\">Optimization workshop<\/a>.  Please join us if interested.  <\/p>\n<p>I always feel like things are going slow, but in the last year, but there have been many changes overall. Notes for <a href=\"https:\/\/groups.yahoo.com\/neo\/groups\/vowpal_wabbit\/conversations\/messages\/3779\">7.7<\/a> are here.  Since then, there are several areas of improvement as well as generalized bug fixes and refactoring.  <\/p>\n<ol>\n<li><b>Learning to Search<\/b>: <a href=\"http:\/\/www.umiacs.umd.edu\/~hal\/\">Hal<\/a> completely rewrote the learning to search system, enough that the numbers <a href=\"http:\/\/arxiv.org\/abs\/1406.1837\">here<\/a> are looking obsolete. <a href=\"http:\/\/web.engr.illinois.edu\/~kchang10\/\">Kai-Wei<\/a> has also created several advanced applications for entity-relation and dependency parsing which are promising.\n<\/li>\n<li><b>Languages<\/b> Hal also created a good python library, which includes call-backs for learning to search.  You can now develop advanced structured prediction solutions in a nice language.  <a href=\"https:\/\/www.linkedin.com\/in\/jmorra\">Jonathan Morra<\/a> also contributed an initial Java interface.<\/li>\n<li><b>Exploration<\/b> The contextual bandit subsystem now allows evaluation of an arbitrary policy, and an exploration library is now factored out into an independent library (principally by <a href=\"http:\/\/research.microsoft.com\/en-us\/people\/lhoang\/default.aspx\">Luong<\/a> with help from <a href=\"http:\/\/www.cs.princeton.edu\/~sssix\/\">Sid<\/a> and <a href=\"http:\/\/www.eecs.berkeley.edu\/~slbird\/GradSite\/Welcome.html\">Sarah<\/a>).  This is critical for real applications because randomization must happen at the point of decision.  <\/li>\n<li><b>Reductions<\/b> The learning reductions subsystem has continued to mature, although the perfectionist in me is still dissatisfied.  As a consequence, it&#8217;s now pretty easy to program new reductions, and the efficiency of these reductions has generally improved.  Several news ones are cooking.<\/li>\n<li><b>Online Learning<\/b> <a href=\"http:\/\/research.microsoft.com\/en-us\/um\/people\/alekha\/\">Alekh<\/a> added an online SVM implementation based on <a href=\"http:\/\/leon.bottou.org\/projects\/lasvm\">LaSVM<\/a>.  This is known to parallelize well via the <a href=\"http:\/\/arxiv.org\/abs\/1310.8243\">para-active approach<\/a>.<\/li>\n<\/ol>\n<p>This project has grown quite a bit&#8212;there are about 30 different people contributing to VW since the last release, and there is now a <a href=\"http:\/\/www.meetup.com\/SF-Vowpal-Wabbit-User-Group\/\">VW meetup<\/a> (December 8th!) in the bay area that I wish I could attend.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>I just created Vowpal Wabbit 7.8, and we are planning to have an increasingly less heretical followup tutorial during the non-&#8220;ski break&#8221; at the NIPS Optimization workshop. Please join us if interested. I always feel like things are going slow, but in the last year, but there have been many changes overall. Notes for 7.7 &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/hunch.net\/?p=222592\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Vowpal Wabbit 7.8 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":[4,42,29,5],"tags":[],"class_list":["post-222592","post","type-post","status-publish","format-standard","hentry","category-announcements","category-code","category-machine-learning","category-teaching"],"_links":{"self":[{"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/222592","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=222592"}],"version-history":[{"count":0,"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/222592\/revisions"}],"wp:attachment":[{"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=222592"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=222592"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=222592"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}