{"id":110322,"date":"2019-05-22T09:24:56","date_gmt":"2019-05-22T01:24:56","guid":{"rendered":"http:\/\/www.fagao.me\/?p=110322"},"modified":"2019-05-22T09:24:56","modified_gmt":"2019-05-22T01:24:56","slug":"%e6%b7%98%e5%ae%9d%e7%94%a8%e5%bc%ba%e5%8c%96%e5%ad%a6%e4%b9%a0%e4%bc%98%e5%8c%96%e5%95%86%e5%93%81%e6%90%9c%e7%b4%a2%e5%90%8e-%e6%80%bb%e6%94%b6%e5%85%a5%e8%83%bd%e6%8f%90%e9%ab%982-%e8%ae%ba-2","status":"publish","type":"post","link":"https:\/\/www.fagao.me\/p\/110322.htm","title":{"rendered":"\u6dd8\u5b9d\u7528\u5f3a\u5316\u5b66\u4e60\u4f18\u5316\u5546\u54c1\u641c\u7d22\u540e \u603b\u6536\u5165\u80fd\u63d0\u9ad82% | \u8bba\u6587"},"content":{"rendered":"<p><img decoding=\"async\" src=\"https:\/\/www.fagao.me\/images\/uploads\/2018_007.jpg\" alt=\"\"><\/p>\n<div class=\"articleend\"><\/div>\n<p>\u6797\u9cde \u7f16\u8bd1\u81ea arXiv <\/p>\n<p>\u773c\u770b618\u5728\u5373\uff0c\u6dd8\u5b9d\u53c8\u516c\u5e03\u4e86\u65b0\u7814\u7a76\u6210\u679c\u3002<\/p>\n<p>\u8fd1\u65e5\uff0c\u5357\u4eac\u5927\u5b66\u548c\u6dd8\u5b9d\u8054\u5408\u53d1\u8868\u7684\u8bba\u6587Virtual-Taobao: Virtualizing Real-world Online Retail Environment for Reinforcement Learning\u4e2d\uff0c\u8be6\u7ec6\u4ecb\u7ecd\u4e86\u6dd8\u5b9d\u7528\u5f3a\u5316\u5b66\u4e60\u4f18\u5316\u5546\u54c1\u641c\u7d22\u7684\u65b0\u6280\u672f\u3002<\/p>\n<p>\u8fd9\u4e2a\u65b0\u6784\u5efa\u7684\u201c\u865a\u62df\u6dd8\u5b9d\u201d\u6a21\u62df\u5668\uff0c\u53ef\u4ee5\u8ba9\u7b97\u6cd5\u4ece\u4e70\u5bb6\u7684\u5386\u53f2\u884c\u4e3a\u4e2d\u5b66\u4e60\uff0c\u89c4\u5212\u6700\u4f73\u5546\u54c1\u641c\u7d22\u663e\u793a\u7b56\u7565\uff0c<strong>\u80fd\u5728\u771f\u5b9e\u73af\u5883\u4e2d\u8ba9\u6dd8\u5b9d\u7684\u6536\u5165\u63d0\u9ad82%<\/strong>\uff0c\u662f\u4e00\u7b14\u4e0d\u5c0f\u7684\u6570\u989d\u3002<\/p>\n<p>\u4e00\u8d77\u770b\u770b\u8fd9\u4e2a\u7814\u7a76\u7a76\u7adf\u8bb2\u4e86\u4ec0\u4e48\u2014\u2014<\/p>\n<p>?<\/p>\n<\/p>\n<p>?<\/p>\n<p>\u201c\u56db\u4f4d\u4e00\u4f53\u201d\u7684\u865a\u62df\u6dd8\u5b9d <\/p>\n<p>\u5f3a\u5316\u5b66\u4e60\uff08Reinforcement Learning\uff0cRL\uff09\u662f\u5339\u9ed1\u9a6c\uff0c\u53ef\u80fd\u4f1a\u5bf9\u6dd8\u5b9d\u7528\u6237\u4f53\u9a8c\u4ea7\u751f\u53d8\u9769\u6027\u5f71\u54cd\uff0c\u4f46\u5728\u7269\u7406\u4e16\u754c\u4e2d\u7684RL\u5e94\u7528\u5374\u5c11\u6709\u4eba\u7814\u7a76\u3002<\/p>\n<p>\u8fd9\u662f\u6709\u539f\u56e0\u7684\u3002\u4e00\u822c\u6765\u8bf4\uff0c\u76ee\u524d\u7684RL\u7b97\u6cd5\u901a\u5e38\u9700\u8981\u4e0e\u73af\u5883\u8fdb\u884c\u5927\u91cf\u4ea4\u4e92\uff0c\u6210\u672c\u9ad8\u6602\u3002<\/p>\n<p>\u5728\u8fd9\u4e2a\u9879\u76ee\u4e2d\uff0c\u6dd8\u5b9d\u7814\u7a76\u4eba\u5458\u901a\u8fc7\u4e70\u5bb6\u7684\u5386\u53f2\u8d2d\u4e70\u8bb0\u5f55\uff0c\u751f\u6210\u4e86\u201c<strong>\u865a\u62df\u6dd8\u5b9d\uff08Virtual Taobao\uff09<\/strong>\u201d\uff1b\u5e73\u53f0\uff0c\u865a\u62df\u7684\u4e70\u5bb6\u53ef\u8fdb\u5165\u6dd8\u5b9d\u89e6\u53d1\u5e73\u53f0<a class=\"tag-links\" href=\"https:\/\/www.fagao.me\/p\/t\/121\">\u641c\u7d22\u5f15\u64ce<\/a>\u3002<\/p>\n<p style=\"text-align: right;\">?<\/p>\n<\/p>\n<p>?<strong>\u25b3<\/strong><span style=\"font-size: 16px;\">\u865a\u62df\u6dd8\u5b9d\u67b6\u6784<\/span><\/p>\n<p>\u5728\u8fd9\u91cc\uff0c\u7814\u7a76\u4eba\u5458\u63d0\u51fa\u7684<strong>GAN-SD\uff08GAN-for-Simulating-Distribution\uff09<\/strong>\u7b97\u6cd5\u6a21\u4eff\u865a\u62df\u4e70\u5bb6\u7684\u64cd\u4f5c\u548c\u641c\u7d22\u8bf7\u6c42\u3002<\/p>\n<p>\u865a\u62df\u7528\u6237\u6709\u4e86\uff0c\u4f46\u8fd8\u8fd8\u6ca1\u6709\u548c\u73af\u5883\u4ea7\u751f\u4ea4\u4e92\u3002\u4e3a\u4e86\u8ba9\u865a\u62df\u5e73\u53f0\u4ea7\u751f\u4ea4\u4e92\u53d8\u6210\u52a8\u6001\u73af\u5883\uff0c\u7814\u7a76\u4eba\u5458\u8fd8\u63d0\u51fa\u4e86<strong>MAIL\u65b9\u6cd5\uff08Multi-agent Adversarial Imitation Learning\uff09<\/strong>\uff0c\u4e5f\u53ef\u4ee5\u79f0\u4e4b\u4e3a\u667a\u80fd\u4f53\u5bf9\u6297\u6a21\u4eff\u5b66\u4e60\u6cd5\u3002MAIL\u540c\u65f6\u5b66\u4e60\u4e70\u5bb6\u89c4\u5219\u548c\u5e73\u53f0\u7684\u89c4\u5219\uff0c\u8bad\u7ec3\u4e70\u5bb6\u548c\u5e73\u53f0\u4ea7\u751f\u66f4\u52a0\u771f\u5b9e\u7684\u4ea4\u4e92\u3002<\/p>\n<p>\u4e0d\u8fc7\u865a\u62df\u7684\u7ec8\u7a76\u662f\u5047\u7684\uff0c\u5f53\u7814\u7a76\u4eba\u5458\u53d1\u73b0\u7b97\u6cd5\u8fc7\u5ea6\u62df\u5408\u865a\u62df\u6dd8\u5b9d\u65f6\uff0c\u610f\u5473\u7740\u53ef\u80fd\u5728\u5b9e\u9645\u60c5\u51b5\u4e0b\u8868\u73b0\u4e0d\u4f73\u3002\u5bf9\u6b64\uff0c\u7814\u7a76\u4eba\u5458\u63d0\u51fa\u4e86<strong>\u52a8\u4f5c\u89c4\u8303\u7ea6\u675fANC\u7b56\u7565\uff08Action Norm Constraint\uff09<\/strong>\uff0c\u53ef\u4ee5\u51cf\u5c11\u8fd9\u79cd\u8fc7\u62df\u5408\u3002<\/p>\n<p>\u8fd9\u6837\uff0c\u201c\u56db\u4f4d\u4e00\u4f53\u201d\u7684\u865a\u62df\u6dd8\u5b9d\u5c31\u5efa\u6210\u4e86\u3002<\/p>\n<p>\u5b9e\u9a8c\u7ed3\u679c <\/p>\n<p>\u601d\u8def\u548c\u6a21\u578b\u642d\u5efa\u5b8c\u6bd5\u540e\uff0c\u662f\u65f6\u5019\u8868\u6f14\u771f\u6b63\u7684\u6280\u672f\u4e86\u3002\u7814\u7a76\u4eba\u5458\u7528\u4e0b\u9762\u4e09\u4e2a\u6307\u6807\u8861\u91cf\u5b9e\u9a8c\u7ed3\u679c\u3002<\/p>\n<ul>\n<li>\u603b\u8425\u4e1a\u989d\uff08TT\uff09\uff1a\u6240\u552e\u5546\u54c1\u7684\u4ef7\u503c\u3002<\/li>\n<li>\u603b\u6210\u4ea4\u91cf\uff08TV\uff09\uff1a\u6240\u552e\u5546\u54c1\u7684\u6570\u91cf\u3002<\/li>\n<li>\u9875\u9762\u91c7\u8d2d\u7387\uff08R2P\uff09\uff1a\u91c7\u8d2d\u53d1\u751f\u65f6PV\uff08\u663e\u793a\u9875\u9762\u89c6\u56fe\uff09\u6570\u91cf\u6bd4\u7387\u3002<\/li>\n<\/ul>\n<p>\u5b9e\u9a8c\u7ed3\u679c\u663e\u793a\uff0c\u901a\u8fc7\u6570\u4ebf\u7528\u6237\u8bb0\u5f55\u6784\u5efa\u7684\u865a\u62df\u6dd8\u5b9d\u91cd\u5efa\u4e86\u975e\u5e38\u63a5\u8fd1\u771f\u5b9e\u7269\u7406\u4e16\u754c\u7684\u73af\u5883\u3002<\/p>\n<p style=\"text-align: right;\">?<\/p>\n<\/p>\n<p>?<strong>\u25b3<\/strong><span style=\"font-size: 16px;\">\u6dd8\u5b9d\u548c\u865a\u62df\u6dd8\u5b9d\u95f4\u7684\u7528\u6237\u5206\u5e03\u60c5\u51b5<\/span><\/p>\n<p style=\"text-align: right;\">?<\/p>\n<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/www.fagao.me\/images\/uploads\/2019_018.jpg\" alt=\"\"><\/p>\n<p>?<strong>\u25b3<\/strong><span style=\"font-size: 16px;\">\u6dd8\u5b9d\u548c\u865a\u62df\u6dd8\u5b9d\u4e4b\u95f4\u7684R2P\u5206\u5e03\u60c5\u51b5<\/span><\/p>\n<p>\u7814\u7a76\u4eba\u5458\uff0c\u5c06\u4e00\u5929\u7684\u5386\u53f2\u6570\u636e\u6309\u65f6\u95f4\u987a\u5e8f\u5206\u4e3a12\u4e2a\u90e8\u5206\uff0c\u4ee5\u6a21\u62dfR2P\u968f\u65f6\u95f4\u53d8\u5316\u7684\u8fc7\u7a0b\uff0c\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<p style=\"text-align: right;\">?<\/p>\n<\/p>\n<p>?<strong>\u25b3<\/strong><span style=\"font-size: 16px;\">R2P\u968f\u65f6\u95f4\u63a8\u79fb\u7684\u8d70\u52bf<\/span><\/p>\n<p>\u7814\u7a76\u4eba\u5458\u7528\u4e86\u89c2\u5bdf\u4e86\u865a\u62df\u6dd8\u5b9d\u7684\u6cdb\u5316\u80fd\u529b\uff0c\u5e76\u4e14\u89c2\u5bdf\u7528\u884c\u4e3a\u514b\u9686\u6cd5\uff08BC\uff09\u66ff\u4ee3MAIL\u4e4b\u540eR2P\u8d70\u52bf\u3002\u5b9e\u9a8c\u8bc1\u660eBC\u73af\u5883\u4e2d\u7684R2P\u4e0b\u964d\u66f4\u5feb\u3002<\/p>\n<p style=\"text-align: right;\">?<\/p>\n<\/p>\n<p>?<strong>\u25b3<\/strong><span style=\"font-size: 16px;\">\u865a\u62df\u6dd8\u5b9d\u7684\u6cdb\u5316\u80fd\u529b<\/span><\/p>\n<p>\u6700\u540e\uff0c\u7814\u7a76\u4eba\u5458\u5c06\u865a\u62df\u6dd8\u5b9d\uff08RL + VTaobao\uff09\u4e2d\u7684RL\u65b9\u6cd5\u4ea7\u751f\u7684\u7b56\u7565\u4e0e\u5386\u53f2\u6570\u636e\uff08SL + Data\uff09\u4e0a\u7684\u76d1\u7763\u5f0f\u5b66\u4e60\u65b9\u6cd5\u8fdb\u884c\u6bd4\u8f83\uff0c\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<p>?<\/p>\n<\/p>\n<p>?<\/p>\n<p>\u53ef\u4ee5\u770b\u51fa\uff0c\u8be5RL+ VTaobao\u603b\u662f\u4f18\u4e8eSL+Data\u6cd5\u3002<\/p>\n<p>\u4f5c\u8005\u56e2\u961f <\/p>\n<p>\u8fd9\u4e00\u7814\u7a76\u7684\u4f5c\u8005\u6709\u4e94\u4eba\uff0c\u5305\u62ec\u5357\u4eac\u5927\u5b66\u8f6f\u4ef6\u65b0\u6280\u672f\u56fd\u5bb6\u91cd\u70b9\u5b9e\u9a8c\u5ba4\u7684Jing Cheng Shi\uff0cYang Yu\uff0cShi Yong Chen\uff0c\u4e5f\u6709\u963f\u91cc\u96c6\u56e2\u7684Qing Da\u548c\u66fe\u5b89\u7965\uff08\u82b1\u540d\u4ec1\u91cd\uff09\u3002<\/p>\n<p style=\"text-align: center;\">?<\/p>\n<\/p>\n<p>?<\/p>\n<p><strong>\u25b3<\/strong>\u963f\u91cc\u5df4\u5df4\u641c\u7d22\u4e8b\u4e1a\u90e8\u66fe\u5b89\u7965 <\/p>\n<p>\u963f\u91cc\u96c6\u56e2\u548c\u5357\u4eac\u5927\u5b66\u7684\u5408\u4f5c\u4e0d\u53ea\u6dd8\u5b9d\u8fd9\u4e2a\u6848\u4f8b\u3002\u4e0d\u4e45\u524d\uff0c\u5b9e\u9a8c\u5ba4\u4e0e\u8682\u8681\u91d1\u670d\u5408\u8457\u7684\u7814\u7a76Distributed Deep Forest and its Application to Automatic Detection of Cash-out Fraud\u8bba\u6587\u4e5f\u5df2\u53d1\u8868\u3002\u5bf9\u4e86\uff0c\u8fd9\u4e2a\u5b9e\u9a8c\u5ba4\u7684\u5e38\u52a1\u526f\u4e3b\u4efb\u662f\u5468\u5fd7\u534e\u3002<\/p>\n<p>\u76f8\u5173\u8d44\u6599 <\/p>\n<p>Virtual-Taobao\u8bba\u6587\u5730\u5740\uff1a<\/p>\n<p>https:\/\/arxiv.org\/abs\/1805.10000<\/p>\n<p>\u8682\u8681\u91d1\u670d\u00d7\u5357\u4eac\u5927\u5b66\u8bba\u6587\u5730\u5740\uff1a<\/p>\n<p>https:\/\/arxiv.org\/abs\/1805.04234<\/p>\n<p>\u2014 <strong>\u5b8c<\/strong>\u2014<\/p>\n<div class=\"articleend\"><\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u6797\u9cde \u7f16\u8bd1\u81ea arXiv \u773c\u770b618\u5728\u5373\uff0c\u6dd8\u5b9d\u53c8\u516c\u5e03\u4e86\u65b0\u7814\u7a76\u6210\u679c\u3002 \u8fd1\u65e5\uff0c\u5357\u4eac\u5927\u5b66\u548c\u6dd8\u5b9d\u8054\u5408\u53d1\u8868\u7684\u8bba\u6587Virt [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[183],"class_list":["post-110322","post","type-post","status-publish","format-standard","hentry","category-news","tag-183"],"_links":{"self":[{"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/posts\/110322","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/comments?post=110322"}],"version-history":[{"count":0,"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/posts\/110322\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/media?parent=110322"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/categories?post=110322"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.fagao.me\/p\/wp-json\/wp\/v2\/tags?post=110322"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}