{"id":35,"date":"2005-02-27T11:16:28","date_gmt":"2005-02-27T17:16:28","guid":{"rendered":"\/?p=35"},"modified":"2005-02-28T13:19:49","modified_gmt":"2005-02-28T19:19:49","slug":"antilearning-proximity-is-bad","status":"publish","type":"post","link":"https:\/\/hunch.net\/?p=35","title":{"rendered":"Antilearning: When proximity goes bad"},"content":{"rendered":"<p><a href=\"http:\/\/www.princeton.edu\/~jpredd\/\">Joel Predd<\/a> <a href=\"https:\/\/hunch.net\/index.php?p=22\">mentioned<\/a> &#8220;<a href=\"http:\/\/wwwmaths.anu.edu.au\/events\/sy2005\/programs.html#Kowalczyk\">Antilearning<\/a>&#8221; by <a href=\"http:\/\/nicta.com.au\/director\/research\/programs\/sml\/people\/adam_kowalczyk.cfm\">Adam Kowalczyk<\/a>, which is interesting from a foundational intuitions viewpoint.<\/p>\n<p>There is a pervasive intuition that &#8220;nearby things tend to have the same label&#8221;.  This intuition is instantiated in SVMs, nearest neighbor classifiers, decision trees, and neural networks.  It turns out there are natural problems where this intuition is opposite of the truth.<\/p>\n<p>One natural situation where this occurs is in competition.   For example, when <a href=\"http:\/\/intel.com\">Intel<\/a> fails to meet its earnings estimate, is this evidence that <a href=\"http:\/\/www.amd.com\/us-en\/\">AMD<\/a> is doing badly also?  Or evidence that AMD is doing well?<\/p>\n<p>This violation of the proximity intuition means that when the number of examples is few, <em>negating<\/em> a classifier which attempts to exploit proximity can provide predictive power (thus, the term &#8220;antilearning&#8221;).<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Joel Predd mentioned &#8220;Antilearning&#8221; by Adam Kowalczyk, which is interesting from a foundational intuitions viewpoint. There is a pervasive intuition that &#8220;nearby things tend to have the same label&#8221;. This intuition is instantiated in SVMs, nearest neighbor classifiers, decision trees, and neural networks. It turns out there are natural problems where this intuition is opposite &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/hunch.net\/?p=35\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Antilearning: When proximity goes bad&#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":[1],"tags":[],"class_list":["post-35","post","type-post","status-publish","format-standard","hentry","category-general"],"_links":{"self":[{"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/35","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=35"}],"version-history":[{"count":0,"href":"https:\/\/hunch.net\/index.php?rest_route=\/wp\/v2\/posts\/35\/revisions"}],"wp:attachment":[{"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=35"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=35"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hunch.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=35"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}