{"id":24139,"date":"2019-04-12T20:13:02","date_gmt":"2019-04-13T04:13:02","guid":{"rendered":"http:\/\/philiplelyveld.com\/?p=24139"},"modified":"2019-04-12T20:13:02","modified_gmt":"2019-04-13T04:13:02","slug":"a-leading-ai-conference-is-trying-to-fix-the-fields-reproducibility-crisis","status":"publish","type":"post","link":"http:\/\/philiplelyveld.com\/?p=24139","title":{"rendered":"A leading AI conference is trying to fix the field\u2019s reproducibility crisis"},"content":{"rendered":"<p><a href=\"http:\/\/philiplelyveld.com\/wordpress\/wp-content\/uploads\/2019\/04\/reproducibility.png\"><img loading=\"lazy\" decoding=\"async\" class=\"alignleft size-medium wp-image-24140\" src=\"http:\/\/philiplelyveld.com\/wordpress\/wp-content\/uploads\/2019\/04\/reproducibility-300x169.png\" alt=\"reproducibility\" width=\"300\" height=\"169\" srcset=\"http:\/\/philiplelyveld.com\/wordpress\/wp-content\/uploads\/2019\/04\/reproducibility-300x169.png 300w, http:\/\/philiplelyveld.com\/wordpress\/wp-content\/uploads\/2019\/04\/reproducibility-768x432.png 768w, http:\/\/philiplelyveld.com\/wordpress\/wp-content\/uploads\/2019\/04\/reproducibility-1024x575.png 1024w, http:\/\/philiplelyveld.com\/wordpress\/wp-content\/uploads\/2019\/04\/reproducibility.png 2000w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a>Last week, organizers of the Neural Information Processing Systems Conference (NeurIPS), one of the world\u2019s largest annual AI research conferences, <a href=\"https:\/\/medium.com\/@NeurIPSConf\/call-for-papers-689294418f43\">updated<\/a>\u00a0their policy for paper submissions to require what they\u2019re calling a reproducibility checklist. It\u2019s a small shift in a grander fight to curb the growing \u201creproducibility crisis\u201d in science, where a disconcerting number of research findings are not successfully being replicated by other researchers, casting doubt on the validity of the initial findings.<\/p>\n<p>In February, a statistician from Rice University <a href=\"https:\/\/www.technologyreview.com\/the-download\/612982\/machine-learning-is-contributing-to-a-reproducibility-crisis-within-science\/\">warned<\/a> that machine-learning techniques are likely fueling that crisis because the results they produce are difficult to audit. It\u2019s a worrying problem as machine learning is increasingly being applied in important areas such as health care and drug research.<\/p>\n<p>As the \u201cworld's most significant AI conference,\u201d wrote Jack Clark, the policy director of the nonprofit OpenAI, in his weekly newsletter Import AI, \u201cNeurIPS 2019 policy will have [a] knock-on effect across [the] wider AI ecosystem.\u201d<\/p>\n<p>See the full story <a href=\"https:\/\/www.technologyreview.com\/the-download\/613297\/the-biggest-ai-conference-is-trying-to-fix-the-fields-reproducibility-crisis\/?fbclid=IwAR0FVhzvXuqUaRZMng5uEXDdG45AYuxFPQoUOQ0RdIAKv92BK5z_l1SXnOs\">here:\u00a0https:\/\/www.technologyreview.com\/the-download\/613297\/the-biggest-ai-conference-is-trying-to-fix-the-fields-reproducibility-crisis\/?fbclid=IwAR0FVhzvXuqUaRZMng5uEXDdG45AYuxFPQoUOQ0RdIAKv92BK5z_l1SXnOs<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Last week, organizers of the Neural Information Processing Systems Conference (NeurIPS), one of the world\u2019s largest annual AI research conferences, updated\u00a0their policy for paper submissions to require what they\u2019re calling a reproducibility checklist. It\u2019s a small shift in a grander fight to curb the growing \u201creproducibility crisis\u201d in science, where a disconcerting number of research [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-24139","post","type-post","status-publish","format-standard","hentry","category-non-3d-stories"],"_links":{"self":[{"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=\/wp\/v2\/posts\/24139","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=24139"}],"version-history":[{"count":1,"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=\/wp\/v2\/posts\/24139\/revisions"}],"predecessor-version":[{"id":24141,"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=\/wp\/v2\/posts\/24139\/revisions\/24141"}],"wp:attachment":[{"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=24139"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=24139"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/philiplelyveld.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=24139"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}