{"id":5937325,"date":"2017-01-04T14:24:08","date_gmt":"2017-01-04T20:24:08","guid":{"rendered":"http:\/\/hunch.net\/?p=5937325"},"modified":"2017-01-13T18:14:29","modified_gmt":"2017-01-14T00:14:29","slug":"ewrl-and-nips-2016","status":"publish","type":"post","link":"https:\/\/hunch.net\/?p=5937325","title":{"rendered":"EWRL and NIPS 2016"},"content":{"rendered":"

I went to the European Workshop on Reinforcement Learning<\/a> and NIPS<\/a> last month and saw several interesting things. <\/p>\n

At EWRL, I particularly liked the talks from: <\/p>\n

    \n
  1. Remi Munos<\/a> on off-policy evaluation<\/li>\n
  2. Mohammad Ghavamzadeh<\/a> on learning safe policies<\/li>\n
  3. Emma Brunskill on optimizing biased-but safe estimators (sense a theme?)<\/li>\n
  4. Sergey Levine<\/a> on low sample complexity applications of RL in robotics.<\/li>\n<\/ol>\n

    My talk is here<\/a>. Overall, this was a well organized workshop with diverse and interesting subjects, with the only caveat being that they had to limit registration \ud83d\ude42<\/p>\n

    At NIPS itself, I found the poster sessions fairly interesting. <\/p>\n

      \n
    1. Allen-Zhu and Hazan had a new notion of a reduction<\/a> (video<\/a>).<\/li>\n
    2. Zhao, Poupart, and Gordon had a new way to learn Sum-Product Networks<\/a><\/li>\n
    3. Ho, Littman, MacGlashan, Cushman, and Austerwell, had a paper on how “Showing” is different from “Doing”<\/a>.<\/li>\n
    4. Toulis and Parkes had a paper on estimation of long term causal effects<\/a>.<\/li>\n
    5. Rae, Hunt, Danihelka, Harley, Senior, Wayne, Graves, and Lillicrap had a paper on large memories with neural networks<\/a>.\n<\/li>\n
    6. Hardt, Price, and Srebro, had a paper on Equal Opportunity in ML<\/a>.<\/li>\n<\/ol>\n

      Format-wise, I thought the 2 sessions was better than 1, but I really would have preferred more. The recorded spotlights are also pretty cool.<\/p>\n

      The NIPS workshops were great, although I was somewhat reminded of kindergarten soccer<\/a> in terms of lopsided attendance. This may be inevitable given how hot the field is, but I think it’s important for individual researchers to remember that:<\/p>\n

        \n
      1. There are many important directions of research.<\/li>\n
      2. You personally have a much higher chance of doing something interesting if everyone else is not doing it also.<\/li>\n<\/ol>\n

        During the workshops, I learned about ADAM<\/a> (a momentum form of Adagrad), testing ML systems<\/a>, and that even TenserFlow<\/a> is finally looking into synchronous updates for parallel learning (allreduce<\/a> is the way).<\/p>\n

        (edit: added one)<\/p>\n","protected":false},"excerpt":{"rendered":"

        I went to the European Workshop on Reinforcement Learning and NIPS last month and saw several interesting things. At EWRL, I particularly liked the talks from: Remi Munos on off-policy evaluation Mohammad Ghavamzadeh on learning safe policies Emma Brunskill on optimizing biased-but safe estimators (sense a theme?) Sergey Levine on low sample complexity applications of … <\/p>\n

        Continue reading “EWRL and NIPS 2016”<\/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":[33,29,11],"tags":[],"class_list":["post-5937325","post","type-post","status-publish","format-standard","hentry","category-conferences","category-machine-learning","category-reinforcement"],"_links":{"self":[{"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/5937325","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=5937325"}],"version-history":[{"count":0,"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/5937325\/revisions"}],"wp:attachment":[{"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5937325"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5937325"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5937325"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}