{"id":32061,"date":"2022-01-24T19:53:36","date_gmt":"2022-01-24T11:53:36","guid":{"rendered":"https:\/\/swarma.org\/?p=32061"},"modified":"2022-01-24T19:53:36","modified_gmt":"2022-01-24T11:53:36","slug":"%e5%a4%8d%e6%9d%82%e7%b3%bb%e7%bb%9f%e7%9a%84%e8%b7%a8%e5%ad%a6%e7%a7%91%e6%8e%a2%e7%b4%a2-%e5%8c%97%e4%ba%ac%e5%a4%a7%e5%ad%a6ai%e7%a0%94%e7%a9%b6%e9%99%a2%e5%a4%9a%e6%99%ba%e8%83%bd%e4%bd%93","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=32061","title":{"rendered":"\u590d\u6742\u7cfb\u7edf\u7684\u8de8\u5b66\u79d1\u63a2\u7d22 | \u5317\u4eac\u5927\u5b66AI\u7814\u7a76\u9662\u591a\u667a\u80fd\u4f53\u4e2d\u5fc3\u5b66\u672f\u51ac\u4ee4\u8425\u62db\u52df"},"content":{"rendered":"
\n
\n
\n
\n
\n
\n
\n

<\/p>\n<\/section>\n

\n
\n
\n
\n
\n
\n

<\/section>\n<\/section>\n
\n
\n

\u5bfc\u8bed<\/strong><\/p>\n<\/section>\n<\/section>\n

\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n

<\/strong>\u5317\u4eac\u5927\u5b66\u4eba\u5de5\u667a\u80fd\u7814\u7a76\u9662\u591a\u667a\u80fd\u4f53\u4e2d\u5fc3\u5b66\u672f\u51ac\u4ee4\u8425\u5f00\u59cb\u62db\u52df\u5566\uff5e<\/strong><\/p>\n


<\/strong><\/p>\n

\u5728\u8fd9\u91cc\uff0c\u4f60\u4e0d\u4ec5\u53ef\u4ee5\u8ddf\u968f\u4e2d\u5fc3\u7684\u8001\u5e08\u4eec\u8fdb\u884c\u9879\u76ee\u7684\u7814\u7a76\u4e0e\u5b66\u4e60\uff0c\u589e\u52a0\u4f60\u7684\u79d1\u7814\u5b9e\u8df5\u7ecf\u5386\uff1b\u66f4\u80fd\u4e30\u5bcc\u4eba\u751f\u4f53\u9a8c\uff0c\u63d0\u9ad8\u4f60\u7684\u72ec\u7acb\u81ea\u4e3b\u80fd\u529b\u548c\u521b\u65b0\u7cbe\u795e\u3002<\/strong><\/p>\n


<\/strong><\/p>\n

2022~\u5c31\u8ba9\u5317\u4eac\u5927\u5b66\u4eba\u5de5\u667a\u80fd\u7814\u7a76\u9662\u591a\u667a\u80fd\u4f53\u4e2d\u5fc3\u5b66\u672f\u51ac\u4ee4\u8425\u966a\u4f34\u4f60\u5f00\u542f\u65b0\u7684\u4e00\u5e74\u5427\uff5e<\/strong><\/p>\n


<\/p>\n

\n
\n
\n
\n

\u5317\u4eac\u5927\u5b66\u4eba\u5de5\u667a\u80fd\u7814\u7a76\u9662
<\/span><\/strong><\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n
\n

WINTER CAMP<\/span><\/strong><\/span><\/p>\n<\/section>\n<\/section>\n

\n
\n

\u5b66\u672f\u51ac\u4ee4\u8425<\/span><\/strong><\/span><\/p>\n<\/section>\n<\/section>\n

\n

\u65f6\u95f4\uff1a1\u670815\u65e5-2\u670821\u65e5<\/span><\/strong><\/span><\/p>\n

\u9879\u76ee\u9884\u8ba1\u622a\u6b62\u65e5\u671f\uff1a5\u670831\u65e5<\/span><\/strong><\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n
\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
<\/section>\n<\/section>\n

<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

\u9879\u76ee\u65b9\u5411<\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n<\/section>\n

\u57fa\u4e8e\u5408\u4f5c\u578b\u591a\u667a\u80fd\u4f53\u5f3a\u5316\u5b66\u4e60\u6280\u672f\u7684\u53cc\u7075\u5de7\u624b\u63a7\u5236<\/span><\/strong><\/span>
<\/section>\n

<\/span><\/strong><\/span><\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u8001\u5e08\uff1a\u6768\u8000\u4e1c

\u8bfe\u9898\uff1a\u8be5\u8bfe\u9898\u5c06\u63a2\u7d22\u5982\u4f55\u63a7\u5236\u4e24\u53ea\u7075\u5de7\u624b\u8fdb\u884c\u7684\u5408\u4f5c\u4efb\u52a1\uff0c\u4f8b\u5982\u7aef\u76d8\u5b50\uff0c\u63c9\u5706\u5b50\uff0c\u8f6c\u9b54\u65b9\u7b49\u3002\u6211\u4eec\u5c06\u63a2\u7d22\u5982\u4f55\u4f7f\u7528\u591a\u667a\u80fd\u4f53\u5f3a\u5316\u5b66\u4e60\u6280\u672f\u8fdb\u884c\u5bf9\u53cc\u7075\u5de7\u624b\u63a7\u5236\u7684\u5efa\u6a21\uff0c\u5e76\u4e14\u4ece\u5408\u4f5c\u535a\u5f08\u7684\u89d2\u5ea6\u51fa\u53d1\u8bbe\u8ba1\u7b97\u6cd5\u3002<\/span><\/section>\n

\u53c2\u8003\u6587\u732e\uff1ahttps:\/\/openreview.net\/pdf?id=7uSBJDoP7tY   https:\/\/openreview.net\/pdf?id=EcGGFkNTxdJ

<\/span><\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u719f\u6089\u5f3a\u5316\u5b66\u4e60\u57fa\u672c\u6982\u5ff5\uff0c\u6709\u8c03\u8bd5\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u76f8\u5173\u7ecf\u9a8c\u3002\u7f16\u7a0b\u80fd\u529b\u51fa\u4f17\u53ef\u7834\u683c\u3002

<\/span><\/section>\n
\u5468\u671f\uff1a6-12\u4e2a\u6708\u3002<\/span><\/section>\n<\/section>\n
\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\u5e26\u6709\u5b89\u5168\u7ea6\u675f\u7684\u591a\u667a\u80fd\u4f53\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u8bbe\u8ba1\u4e0e\u5e94\u7528<\/span>
<\/strong><\/span><\/section>\n

<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u8001\u5e08\uff1a\u6768\u8000\u4e1c

\u8bfe\u9898\uff1a\u7b26\u5408\u5b89\u5168\u6027\u7684\u5f3a\u5316\u5b66\u4e60\u6280\u672f\u5177\u6709\u7740\u91cd\u8981\u5e94\u7528\uff0c\u4f8b\u5982\u65e0\u4eba\u9a7e\u9a76\uff0c\u673a\u68b0\u5236\u9020\u7b49\u7b49\u3002\u672c\u8bfe\u9898\u5c06\u7814\u7a76\u5982\u4f55\u5728\u4e00\u4e2a\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u4e2d\u8bbe\u8ba1\u5177\u6709\u5b89\u5168\u6027\u7ea6\u675f\u7684\u5f3a\u5316\u5b66\u4e60\u8fc7\u7a0b\u3002\u6211\u4eec\u5c06\u63a2\u7d22\u6ee1\u8db3\u5b89\u5168\u6027\u7ea6\u675f\u7684\u5f3a\u5316\u5b66\u4e60\u6846\u67b6, \u5e76\u63a2\u7d22\u5982\u4f55\u9488\u5bf9\u4e00\u4e2a\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u8fdb\u884c\u5b89\u5168\u6027\u7684\u5b66\u4e60\u3002

\u53c2\u8003\u6587\u732e\uff1ahttps:\/\/arxiv.org\/abs\/2110.02793

\u6280\u80fd\u8981\u6c42\uff1a\u719f\u6089\u5f3a\u5316\u5b66\u4e60\u57fa\u672c\u6982\u5ff5\uff0c\u6709\u8c03\u8bd5\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u76f8\u5173\u7ecf\u9a8c\u3002\u7f16\u7a0b\u80fd\u529b\u51fa\u4f17\u53ef\u7834\u683c\u3002

\u65f6\u671f\uff1a6-12\u4e2a\u6708\u3002<\/span><\/section>\n

<\/span><\/section>\n
\u9762\u5411\u901a\u7528\u4eba\u5de5\u667a\u80fd\u7684\u5143\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u8bbe\u8ba1<\/span><\/strong><\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/section>\n
\u5408\u4f5c\u8001\u5e08\uff1a\u6768\u8000\u4e1c

\u8bfe\u9898\uff1a\u9762\u5411\u901a\u7528\u4eba\u5de5\u667a\u80fd\uff0c\u667a\u80fd\u4f53\u5f80\u5f80\u9700\u8981\u80fd\u591f\u6709\u540c\u65f6\u5b8c\u6210\u591a\u4e2a\u4efb\u52a1\u7684\u80fd\u529b\u3002\u672c\u8bfe\u9898\u4ece\u6570\u5b66\u5efa\u6a21\u7684\u89d2\u5ea6\uff0c\u5bf9\u5143\u5f3a\u5316\u5b66\u4e60\u5efa\u6a21\u5e76\u4e14\u8bbe\u8ba1\u5b66\u4e60\u7b97\u6cd5\uff0c\u5176\u76ee\u6807\u4f7f\u7528\u5206\u5c42\u5f3a\u5316\u5b66\u4e60\uff0c\u5185\u5c42\u4ee3\u8868\u4e00\u4e2a\u667a\u80fd\u4f53\u7531\u81ea\u8eab\u5956\u52b1\u51fd\u6570\u9a71\u52a8\u7684\u5f3a\u5316\u5b66\u4e60\uff0c\u5916\u5c42\u4ee3\u8868\u8de8\u4efb\u52a1\u7684\u5b66\u4e60\u7ea6\u675f\uff0c\u4ece\u800c\u8fbe\u5230\u8ba9\u667a\u80fd\u4f53\u62e5\u6709\u5143\u5b66\u4e60\u4e0d\u540c\u4efb\u52a1\u7684\u80fd\u529b\u3002

\u53c2\u8003\u6587\u732e\uff1ahttps:\/\/arxiv.org\/abs\/2112.15400

\u6280\u80fd\u8981\u6c42\uff1a\u719f\u6089\u5f3a\u5316\u5b66\u4e60\u57fa\u672c\u6982\u5ff5\uff0c\u6709\u8c03\u8bd5\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u76f8\u5173\u7ecf\u9a8c\u3002\u7f16\u7a0b\u80fd\u529b\u51fa\u4f17\u53ef\u7834\u683c\u3002

\u65f6\u671f\uff1a6-12\u4e2a\u6708\u3002<\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n

<\/section>\n<\/section>\n
\n
\u7fa4\u4f53\u535a\u5f08<\/span><\/strong><\/span>
<\/section>\n

<\/span><\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u6559\u5e08\uff1a\u674e\u963f\u660e<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\uff1a\u4ece\u53ef\u89c2\u5bdf\u5230\u7684\u9e1f\u7fa4\u3001\u9c7c\u7fa4\uff0c\u5230\u4e0d\u6613\u5bdf\u89c9\u7684\u5fae\u751f\u7269\u7fa4\u4f53\uff0c\u518d\u5230\u4f20\u611f\u5668\u3001\u7fa4\u4f53\u673a\u5668\u4eba\u534f\u4f5c\u7cfb\u7edf\u7b49\uff0c\u591a\u7c7b\u81ea\u7136\u4e0e\u4eba\u9020\u7fa4\u4f53\u7cfb\u7edf\u5b58\u5728\u4e8e\u6211\u4eec\u5468\u56f4\u3002\u8fd9\u4e9b\u7531\u591a\u4e2a\u7b80\u5355\u4e2a\u4f53\u7ec4\u6210\u7684\u7fa4\u4f53\u7cfb\u7edf\uff0c\u5c55\u73b0\u4e86\u8d85\u4e4e\u60f3\u8c61\u7684\u8bf8\u591a\u7c7b\u578b\u7684\u667a\u80fd\u884c\u4e3a\u3002\u8fd1\u5e74\u6765\uff0c\u968f\u7740\u4ea4\u53c9\u5b66\u79d1\u7684\u8fc5\u901f\u53d1\u5c55\uff0c\u7fa4\u4f53\u667a\u80fd\u76f8\u5173\u7814\u7a76\u5f97\u5230\u4e86\u4e0d\u540c\u9886\u57df\u79d1\u5b66\u5bb6\u7684\u96c6\u4e2d\u5173\u6ce8\u4e0e\u6df1\u5165\u63a2\u7d22\u3002\u7136\u800c\uff0c\u968f\u7740\u9ad8\u65b0\u6280\u672f\u7684\u98de\u901f\u53d1\u5c55\u548c\u6211\u4eec\u5bf9\u9ad8\u7cbe\u5ea6\u6d77\u91cf\u4e2a\u4f53\u4ea4\u4e92\u4fe1\u606f\u83b7\u53d6\u80fd\u529b\u7684\u4e0d\u65ad\u63d0\u5347\uff0c\u7fa4\u4f53\u7cfb\u7edf\u7684\u7ed3\u6784\u5206\u6790\u3001\u7fa4\u4f53\u4ea4\u4e92\u4e0e\u667a\u80fd\u51b3\u7b56\u8fd9\u4e00\u79d1\u5b66\u95ee\u9898\u9762\u4e34\u91cd\u8981\u6311\u6218\u3002\u672c\u7814\u7a76\u8bfe\u9898\u91cd\u70b9\u5173\u6ce8\u7fa4\u4f53\u7cfb\u7edf\u4e2d\u4e2a\u4f53\u4ea4\u4e92\u535a\u5f08\u3001\u7b56\u7565\u6f14\u5316\u3001\u4e2a\u4f53\u667a\u80fd\u51b3\u7b56\u7b49\u65b9\u9762\u7684\u95ee\u9898\uff0c\u63a2\u7d22\u590d\u6742\u7f51\u7edc\u4e0a\u7fa4\u4f53\u4ea4\u4e92\u535a\u5f08\u3002\u8bfe\u9898\u7ec4\u901a\u8fc7\u5411\u5b66\u751f\u4ecb\u7ecd\u6f14\u5316\u535a\u5f08\u8bba\u3001\u7f51\u7edc\u79d1\u5b66\u7b49\u57fa\u7840\u77e5\u8bc6\u4e3a\u51fa\u53d1\u70b9\uff0c\u8fdb\u4e00\u6b65\u901a\u8fc7\u8f85\u52a9\u9605\u8bfb\u7ecf\u5178\u6587\u732e\u3001\u6fc0\u53d1\u63d0\u51fa\u79d1\u7814\u95ee\u9898\u3001\u63a2\u7d22\u79d1\u7814\u95ee\u9898\u7b49\u65b9\u5f0f\uff0c\u4e3a\u5b66\u751f\u63d0\u4f9b\u63a2\u7d22\u7fa4\u4f53\u667a\u80fd\u76f8\u5173\u95ee\u9898\u7684\u5e73\u53f0\uff0c\u4f18\u79c0\u8005\u53ef\u5728\u77ed\u671f\u79d1\u7814\u8bad\u7ec3\u7ed3\u675f\u540e\u7ee7\u7eed\u53c2\u4e0e\u5230\u8bfe\u9898\u7ec4\u7684\u7814\u7a76\u4e4b\u4e2d\uff0c\u5171\u540c\u53d1\u8868\u5b66\u672f\u53d1\u73b0\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1a<\/span><\/section>\n
Li, A., Zhou, L., Su, Q., Cornelius, S. P., Liu, Y.-Y., Wang, L. & Levin, S. A. Evolution of cooperation on temporal networks. Nature Communications 11, 2259 (2020).<\/span><\/section>\n

<\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u6570\u7406\u57fa\u7840\u624e\u5b9e\uff0c\u638c\u63e1\u57fa\u672c\u7684\u8ba1\u7b97\u673a\u7f16\u7a0b\uff0c\u5bf9\u7406\u8bba\u63a2\u7d22\u6709\u5174\u8da3\u3002<\/span><\/section>\n

<\/section>\n
\u65f6\u671f\uff1a6\u4e2a\u6708\u4ee5\u4e0a<\/span><\/section>\n

<\/section>\n<\/section>\n
\n
\u7f51\u7edc\u80fd\u63a7\u6027<\/span><\/strong><\/span>
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/span><\/section>\n
\u5408\u4f5c\u6559\u5e08: \u674e\u963f\u660e<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\uff1a\u8fd1\u5e74\u6765\uff0c\u7f51\u7edc\u79d1\u5b66\u7684\u8fc5\u901f\u53d1\u5c55\u4e3a\u6211\u4eec\u7406\u89e3\u3001\u5206\u6790\u4e0e\u63a7\u5236\u591a\u673a\u5668\u4eba\u534f\u4f5c\u3001\u667a\u80fd\u7535\u7f51\u3001\u667a\u80fd\u4ea4\u901a\u7b49\u590d\u6742\u7cfb\u7edf\u63d0\u4f9b\u4e86\u6709\u6548\u7684\u5207\u5165\u70b9\u3002\u867d\u7136\u9759\u6001\u7f51\u7edc\u76f8\u5173\u63a2\u7d22\u5df2\u53d6\u5f97\u7a81\u7834\u6027\u8fdb\u5c55\uff0c\u4f46\u662f\u5b9e\u9645\u7cfb\u7edf\u4e2d\u7684\u7f51\u7edc\u5f80\u5f80\u662f\u968f\u7740\u65f6\u95f4\u6f14\u5316\u7684\u65f6\u5e8f\u7f51\u7edc\uff0c\u5373\u7f51\u7edc\u4e2d\u7684\u8282\u70b9\u4e0e\u8fde\u8fb9\u4e0d\u603b\u662f\u5904\u4e8e\u88ab\u6fc0\u6d3b\u7684\u4ea4\u4e92\u72b6\u6001\u3002\u6362\u8a00\u4e4b\uff0c\u7f51\u7edc\u672c\u8eab\u6f14\u5316\u7684\u65f6\u95f4\u5c3a\u5ea6\u5e76\u4e0d\u603b\u662f\u957f\u4e8e\u5176\u4e0a\u7cfb\u7edf\u52a8\u529b\u5b66\u76f8\u5bf9\u5e94\u7684\u65f6\u95f4\u5c3a\u5ea6\u3002\u672c\u8bfe\u9898\u65e8\u5728\u5206\u6790\u65f6\u5e8f\u7f51\u7edc\u5728\u6f14\u5316\u7684\u540c\u65f6\uff0c\u5176\u5bf9\u5e94\u7cfb\u7edf\u80fd\u63a7\u6027\u7684\u53d8\u5316\uff0c\u4ee5\u53ca\u5982\u4f55\u8bbe\u8ba1\u76f8\u5e94\u7684\u5916\u90e8\u63a7\u5236\u8f93\u5165\uff0c\u4f7f\u5f97\u7cfb\u7edf\u72b6\u6001\u6cbf\u7740\u671f\u671b\u65b9\u5411\u6f14\u5316\u3002\u8bfe\u9898\u7ec4\u5c06\u5e26\u9886\u5b66\u751f\u4e86\u89e3\u7f51\u7edc\u63a7\u5236\u8fd9\u4e00\u524d\u6cbf\u9886\u57df\u7684\u76f8\u5173\u57fa\u7840\u77e5\u8bc6\uff0c\u7814\u8bfb\u9886\u57df\u7ecf\u5178\u6587\u732e\uff0c\u63d0\u51fa\u5e76\u63a2\u7a76\u5177\u6709\u521b\u65b0\u6027\u7684\u79d1\u5b66\u95ee\u9898\uff0c\u4eb2\u8eab\u4f53\u4f1a\u65f6\u5e8f\u7f51\u7edc\u63a7\u5236\u5bf9\u4e8e\u5404\u7c7b\u91cd\u5927\u5de5\u7a0b\u6280\u672f\u7cfb\u7edf\u8bbe\u8ba1\u7684\u91cd\u8981\u610f\u4e49\u3002\u4f18\u79c0\u8005\u53ef\u5728\u77ed\u671f\u79d1\u7814\u8bad\u7ec3\u7ed3\u675f\u540e\u7ee7\u7eed\u53c2\u4e0e\u5230\u8bfe\u9898\u7ec4\u7684\u7814\u7a76\u4e4b\u4e2d\uff0c\u5171\u540c\u53d1\u8868\u5b66\u672f\u53d1\u73b0\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1a<\/span><\/section>\n
Li, A., Cornelius, S. P., Liu, Y.-Y., Wang, L. & Barab\u00e1si, A.-L. The fundamental advantages of temporal networks. Science 358, 1042-1046 (2017).<\/span><\/section>\n

<\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u6570\u7406\u57fa\u7840\u624e\u5b9e\uff0c\u638c\u63e1\u57fa\u672c\u7684\u8ba1\u7b97\u673a\u7f16\u7a0b\uff0c\u5bf9\u7406\u8bba\u63a2\u7d22\u6709\u5174\u8da3\u3002<\/span><\/section>\n

<\/section>\n
\u65f6\u671f\uff1a6\u4e2a\u6708\u4ee5\u4e0a<\/span><\/section>\n

<\/section>\n<\/section>\n
\n
\u4e0d\u5b8c\u5168\u4fe1\u606f\u4e8c\u4eba\u535a\u5f08\u7684\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u5206\u6790<\/span><\/strong><\/span>
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/span><\/section>\n
\u5408\u4f5c\u6559\u5e08\uff1a\u9093\u5c0f\u94c1<\/span><\/section>\n

<\/section>\n
\u9898\u76ee\uff1a\u5b9e\u73b0\u4e0d\u5b8c\u5168\u4fe1\u606f\u4e0b\u535a\u5f08\u7684\u7b80\u5355\u7cfb\u7edf\u6846\u67b6\u3001\u6848\u4f8b\u5206\u6790\u3002<\/span><\/section>\n

<\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u575a\u5b9e\u7684\u8ba1\u7b97\u673a\u7f16\u7a0b\u57fa\u7840\uff0c\u6570\u5b66\u57fa\u7840\uff0c\u5bf9\u7406\u8bba\u63a2\u7d22\u611f\u5174\u8da3\u3002<\/span><\/section>\n

<\/section>\n
\u5468\u671f\uff1a\u6309\u4efb\u52a1\u548c\u5de5\u4f5c\u91cf\u5b89\u6392\u4e0d\u540c\uff1a1-3\u4e2a\u6708\u6216\u80052-6\u4e2a\u6708\u6216\u80053-9\u4e2a\u6708\u3002<\/span><\/section>\n<\/section>\n
\n

<\/section>\n<\/section>\n
\n
\u57fa\u4e8e\u673a\u5668\u5b66\u4e60\u7684\u673a\u5236\u8bbe\u8ba1\u7814\u7a76<\/span><\/strong><\/span>
<\/section>\n

<\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u6559\u5e08\uff1a\u9093\u5c0f\u94c1<\/span><\/section>\n

<\/section>\n
\u9898\u76ee\uff1a\u673a\u5236\u8bbe\u8ba1\u5728\u4e92\u8054\u7f51\u5e7f\u544a\u3001\u62cd\u5356\u3001\u653f\u7b56\u5236\u8ba2\u3001\u533a\u5757\u94fe\u4e2d\u626e\u6f14\u7740\u6838\u5fc3\u89d2\u8272\u3002\u6211\u4eec\u62df\u91c7\u7528\u673a\u5668\u5b66\u4e60\u8fdb\u884c\u81ea\u52a8\u673a\u5236\u8bbe\u8ba1\uff0c\u5e76\u8fdb\u884c\u76f8\u5173\u7684\u7406\u8bba\u5206\u6790\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1ahttps:\/\/arxiv.org\/abs\/2110.06880<\/span><\/section>\n

<\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u624e\u5b9e\u7684\u6570\u5b66\u57fa\u7840\uff0c\u4f18\u79c0\u7684\u7f16\u7a0b\u80fd\u529b\uff0c\u6709\u6df1\u5ea6\u5b66\u4e60\u7ecf\u9a8c\u8005\u4f18\u5148\u3002<\/span><\/section>\n

<\/section>\n
\u5468\u671f\uff1a3-9\u4e2a\u6708\uff0c\u4ee5NeurIPS\/ICLR\/AAMAS\u4e3a\u6295\u7a3f\u76ee\u6807\u3002<\/span><\/section>\n

<\/section>\n<\/section>\n
\n
Grounding Language for Generalization in Reinforcement Learning<\/span><\/strong><\/span>
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/section>\n
\u5408\u4f5c\u6559\u5e08\uff1a\u5362\u5b97\u9752<\/span><\/section>\n

<\/section>\n
\u9898\u76ee\uff1a\u6cdb\u5316\u6027\u662f\u5f3a\u5316\u5b66\u4e60\u4e2d\u7684\u91cd\u8981\u95ee\u9898\u4e5f\u662f\u96be\u9898\uff0c\u771f\u6b63\u5b9e\u73b0\u6cdb\u5316\u6027\u9700\u8981\u667a\u80fd\u4f53\u5bf9MDP\u5177\u6709\u62bd\u8c61\u7684\u8868\u793a\u80fd\u529b\uff0c\u76ee\u524d\u6765\u770b\u8fd8\u5f88\u96be\u505a\u5230\u8fd9\u4e00\u70b9\u3002\u4f46\u6211\u4eec\u53ef\u4ee5\u6362\u4e00\u4e2a\u89d2\u5ea6\u601d\u8003\u3002\u4eba\u7c7b\u8bed\u8a00\u662f\u9ad8\u5ea6\u62bd\u8c61\u7684\u8868\u793a\uff0c\u667a\u80fd\u4f53\u80fd\u4e0d\u80fd\u501f\u52a9\u4eba\u7c7b\u8bed\u8a00\u5b9e\u73b0\u7b56\u7565\u6cdb\u5316\u6027\u3002\u672c\u7814\u7a76\u5c06\u63a2\u7d22\u7ed3\u5408\u81ea\u7136\u8bed\u8a00\u5b9e\u73b0\u5f3a\u5316\u5b66\u4e60\u7b56\u7565\u6cdb\u5316\u6027\uff0c\u8ba9\u667a\u80fd\u4f53\u5b66\u4f1agrounding manuals\/instructions to MDPs\uff0c\u4ece\u800c\u5b9e\u73b0\u6cdb\u5316\u6027\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1ahttps:\/\/arxiv.org\/abs\/2101.07393<\/span><\/section>\n

<\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u719f\u6089\u5f3a\u5316\u5b66\u4e60\u7684\u57fa\u672c\u6982\u5ff5\uff0c\u6709\u8c03\u8bd5\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u7684\u7ecf\u9a8c\u3002<\/span><\/section>\n

<\/section>\n
\u5468\u671f: \u5927\u4e8e6\u4e2a\u6708\u3002<\/span><\/section>\n<\/section>\n
\n

<\/section>\n<\/section>\n
\n
\u975e\u5b8c\u7f8e\u4fe1\u606f\u535a\u5f08\u4e2d\u7684\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u8bbe\u8ba1\u4e0e\u5e94\u7528<\/span>
<\/strong><\/span><\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u6559\u5e08\uff1a\u674e\u6587\u65b0<\/span><\/section>\n

<\/section>\n
\u9898\u76ee\uff1a\u8bb8\u591a\u4f20\u7edf\u6e38\u620f\u5c5e\u4e8e\u975e\u5b8c\u7f8e\u4fe1\u606f\u535a\u5f08\uff0c\u5982\u5fb7\u5dde\u6251\u514b\u3001\u9ebb\u5c06\u3001\u6597\u5730\u4e3b\u7b49\uff0c\u5176\u6700\u4f18\u7b56\u7565\u9700\u8981\u8003\u8651\u5230\u975e\u5b8c\u7f8e\u4fe1\u606f\u5e26\u6765\u7684\u4e0d\u786e\u5b9a\u6027\u3002\u6211\u4eec\u8c03\u7814\u8be5\u9886\u57df\u7684\u7814\u7a76\u73b0\u72b6\uff0c\u5e76\u63a2\u7d22\u5982\u4f55\u5c06\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u5e94\u7528\u5230\u7279\u5b9a\u6e38\u620f\u73af\u5883\u4e2d\uff0c\u51cf\u5c11\u968f\u673a\u6027\u5e26\u6765\u7684\u8bad\u7ec3\u4e0d\u7a33\u5b9a\u95ee\u9898\uff0c\u8bad\u51fa\u9ad8\u6c34\u5e73\u7684\u667a\u80fd\u4f53\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1a<\/span><\/section>\n

DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning<\/span><\/p>\n

AlphaHoldem: High-Performance Artificial Intelligence for Heads-Up No-Limit Poker via End-to-End Reinforcement Learning<\/span><\/p>\n


<\/section>\n
\u6280\u80fd\u8981\u6c42\uff1a\u719f\u6089\u5f3a\u5316\u5b66\u4e60\u7684\u57fa\u672c\u539f\u7406\uff0c\u6709\u8c03\u8bd5\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u7684\u7ecf\u9a8c\uff0c\u5bf9\u6e38\u620fAI\u7814\u7a76\u611f\u5174\u8da3\u3002<\/span><\/section>\n

<\/span><\/section>\n
\u57fa\u4e8e\u98ce\u9669\u5ea6\u91cf\u7684\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u7814\u7a76<\/span><\/strong><\/span>
<\/section>\n

<\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u6559\u5e08\uff1a\u5f6d\u4e00\u6770<\/span><\/section>\n

<\/section>\n
\u7814\u7a76\u5185\u5bb9\uff1a\u7ecf\u5178\u7684\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u901a\u5e38\u662f\u56f4\u7ed5\u671f\u671b\u51c6\u5219\u8bbe\u8ba1\u7684\uff0c\u667a\u80fd\u4f53\u5728\u4e0e\u73af\u5883\u7684\u4ea4\u4e92\u4e2d\u8ffd\u6c42\u671f\u671b\u56de\u62a5\u7684\u6700\u5927\u5316\u3002\u671f\u671b\u53cd\u6620\u4e86\u667a\u80fd\u4f53\u8868\u73b0\u7684\u5e73\u5747\u6c34\u5e73\uff0c\u800c\u5bf9\u5176\u5206\u5e03\u5c3e\u90e8\u7684\u6781\u7aef\u8868\u73b0\u6ca1\u6709\u7ea6\u675f\u3002\u98ce\u9669\u5ea6\u91cf\u662f\u4e00\u65cf\u8003\u8651\u4e86\u7cfb\u7edf\u5728\u6781\u7aef\u73af\u5883\u4e0b\u8868\u73b0\u6c34\u5e73\u7684\u5ea6\u91cf\u5de5\u5177\uff0c\u5728\u91d1\u878d\u6295\u8d44\u7b49\u9886\u57df\u7684\u98ce\u9669\u7ba1\u7406\u4e2d\u5df2\u6709\u5e7f\u6cdb\u7684\u5e94\u7528\u3002\u672c\u8bfe\u9898\u5c1d\u8bd5\u5c06\u4e0d\u540c\u7684\u98ce\u9669\u5ea6\u91cf\u878d\u5165\u667a\u80fd\u4f53\u7684\u51b3\u7b56\u4e2d\uff0c\u63d0\u51fa\u56f4\u7ed5\u98ce\u9669\u51c6\u5219\u8fed\u4ee3\u7684\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff0c\u5e76\u5411\u66f4\u590d\u6742\u7684\u591a\u667a\u80fd\u4f53\u8bad\u7ec3\u573a\u666f\u62d3\u5c55\u3002\u4e0d\u540c\u7684\u98ce\u9669\u5ea6\u91cf\u5c06\u653e\u5927\u4eba\u4eec\u6240\u611f\u5174\u8da3\u7684\u6781\u7aef\u8868\u73b0\u7684\u91cd\u8981\u6027\uff0c\u4f7f\u5f97\u667a\u80fd\u4f53\u7684\u603b\u4f53\u8868\u73b0\u66f4\u4e3a\u7a33\u5065\u3002<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\u8981\u6c42\uff1a\u719f\u6089\u7ecf\u5178\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u7406\u8bba\uff1b\u719f\u6089Python\u8bed\u8a00\u53caPytorch\u6846\u67b6\u7684\u4f7f\u7528\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e<\/span><\/section>\n

[1] Chow Y, Ghavamzadeh M, Janson L, et al. Risk-constrained reinforcement learning with percentile risk criteria[J]. The Journal of Machine Learning Research, 2017, 18(1): 6070-6120.<\/span><\/p>\n

[2] Prashanth L A, Jie C, Fu M, et al. Cumulative prospect theory meets reinforcement learning: Prediction and control[C]\/\/International Conference on Machine Learning. PMLR, 2016: 1406-1415.<\/span><\/p>\n


<\/section>\n
\u5468\u671f\uff1a6-12\u4e2a\u6708\u3002<\/span><\/section>\n<\/section>\n
\n

<\/span><\/section>\n
\u57fa\u4e8e\u4f3c\u7136\u6bd4\u65b9\u6cd5\u7684\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u7814\u7a76<\/span><\/strong><\/span>
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/section>\n
\u5408\u4f5c\u6559\u5e08\uff1a\u5f6d\u4e00\u6770<\/span><\/section>\n

<\/section>\n
\u7814\u7a76\u5185\u5bb9\uff1a\u5728\u73b0\u4ee3\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u4e2d\u795e\u7ecf\u7f51\u7edc\u5145\u5f53\u7740\u91cd\u8981\u7684\u89d2\u8272\uff0c\u5bf9\u5176\u68af\u5ea6\u4f30\u8ba1\u65b9\u6cd5\u7684\u6539\u8fdb\u5c06\u5e7f\u6cdb\u5f71\u54cd\u673a\u5668\u5b66\u4e60\u7684\u5e94\u7528\u53d1\u5c55\u3002\u76ee\u524d\u4e3b\u6d41\u7684\u53cd\u5411\u4f20\u64ad\u65b9\u6cd5\u4f9d\u8d56\u4e8e\u8ba1\u7b97\u56fe\u548c\u94fe\u5f0f\u6c42\u5bfc\u6cd5\u5219\uff0c\u5176\u5bf9\u8ba1\u7b97\u56fe\u7684\u8fde\u7eed\u6027\u548c\u5df2\u77e5\u6027\u8981\u6c42\u7ea6\u675f\u4e86\u7f51\u7edc\u5f62\u6001\u7684\u8bbe\u8ba1\uff0c\u4e5f\u8feb\u4f7f\u68af\u5ea6\u4f30\u8ba1\u53ea\u80fd\u505a\u5230\u795e\u7ecf\u5c42\u5185\u7684\u5e76\u884c\u3002\u672c\u8bfe\u9898\u5c1d\u8bd5\u5c06\u4f3c\u7136\u6bd4\u6280\u5de7\u5e94\u7528\u5230\u795e\u7ecf\u7f51\u7edc\u7684\u68af\u5ea6\u4f30\u8ba1\u4e2d\uff0c\u63d0\u51fa\u4e0d\u53d7\u524d\u8ff0\u4f9d\u8d56\u5236\u7ea6\u7684\u4f3c\u7136\u6bd4\u68af\u5ea6\u4f30\u8ba1\u65b9\u6cd5\u3002\u4f3c\u7136\u6bd4\u65b9\u6cd5\u53ef\u4ee5\u5904\u7406\u4e0d\u8fde\u7eed\u7684\u751a\u81f3\u672a\u77e5\u7684\u8ba1\u7b97\u56fe\uff0c\u5728\u4ee5\u8109\u51b2\u4fe1\u53f7\u4e3a\u8f93\u5165\u7684\u65b0\u4e00\u4ee3\u795e\u7ecf\u7f51\u7edc\u4e2d\u6709\u7740\u72ec\u7279\u7684\u5e94\u7528\u524d\u666f\uff1b\u540c\u65f6\uff0c\u8be5\u65b9\u6cd5\u7406\u8bba\u4e0a\u8fd8\u5177\u6709\u5728\u6574\u4e2a\u795e\u7ecf\u7f51\u7edc\u4e2d\u5e76\u884c\u7684\u6f5c\u529b\u3002<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\u8981\u6c42\uff1a\u719f\u6089\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u7684\u57fa\u672c\u7406\u8bba\uff1b\u719f\u6089Python\u6216C++\u8bed\u8a00\u7684\u4f7f\u7528\uff1b\u5bf9\u5e76\u884c\u5316\u7f16\u7a0b\u6709\u7ecf\u9a8c\u8005\u4f18\u5148\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e<\/span><\/section>\n
[1] Peng Y, Xiao L, Heidergott B, et al. A new likelihood ratio method for training artificial neural networks[J]. INFORMS Journal on Computing, 2021.<\/span><\/section>\n
[2] Tavanaei A, Ghodrati M, Kheradpisheh S R, et al. Deep learning in spiking neural networks[J]. Neural Networks, 2019, 111: 47-63.<\/span><\/section>\n

<\/section>\n<\/section>\n
\n
\u57fa\u4e8e\u52a8\u6001\u4eff\u771f\u62bd\u6837\u7684\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\u7b97\u6cd5\u7406\u8bba\u4e0e\u5e94\u7528<\/span><\/strong><\/span>
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/section>\n
\u5408\u4f5c\u6559\u5e08\uff1a\u5f6d\u4e00\u6770<\/span><\/section>\n

<\/section>\n
\u7814\u7a76\u5185\u5bb9\uff1a\u4f20\u7edf\u4e0a\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\u91c7\u7528\u673a\u5668\u5b66\u4e60\u9886\u57df\u7684\u591a\u81c2\u8001\u864e\u673a\u95ee\u9898\u4e2d\u7684UCT\u7b97\u6cd5\u505a\u8282\u70b9\u9009\u62e9\uff0c\u4f46\u8be5\u95ee\u9898\u7684\u5047\u8bbe\u4e0e\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\u7684\u7ed3\u6784\u5e76\u4e0d\u543b\u5408\u3002\u591a\u81c2\u8001\u864e\u673a\u95ee\u9898\u4e2d\u6bcf\u4e00\u6b65\u90fd\u53ef\u4ee5\u83b7\u5f97\u56de\u62a5\uff0c\u800c\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\u4e2d\u9700\u8981\u5230\u7ec8\u6b62\u8282\u70b9\u624d\u80fd\u83b7\u5f97\u56de\u62a5\uff1b\u6b64\u5916\uff0cUCT\u7b97\u6cd5\u4ec5\u5229\u7528\u4e86\u8282\u70b9\u4f30\u8ba1\u7684\u5747\u503c\u4e0e\u8282\u70b9\u88ab\u8bbf\u95ee\u7684\u6b21\u6570\uff0c\u800c\u672a\u5229\u7528\u5982\u8282\u70b9\u4f30\u8ba1\u7684\u65b9\u5dee\u7b49\u5176\u5b83\u4fe1\u606f\u3002\u8bfe\u9898\u5c1d\u8bd5\u5c06\u4eff\u771f\u4f18\u5316\u9886\u57df\u7684\u7edf\u8ba1\u6392\u5e8f\u4e0e\u9009\u62e9\u95ee\u9898\u7684\u7b56\u7565\u878d\u5165\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\u4e2d\uff0c\u63d0\u51fa\u91c7\u7528\u52a8\u6001\u4eff\u771f\u8d44\u6e90\u5206\u914d\u7b56\u7565\u505a\u8282\u70b9\u9009\u62e9\u7684\u9ad8\u6548\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\u7b97\u6cd5\uff0c\u5e76\u5728\u5bf9\u5f08\u6e38\u620f\u4e2d\u5e94\u7528\uff1b\u5f00\u53d1\u5bf9\u5f08\u6e38\u620f\u5e73\u53f0\uff0c\u4fbf\u4e8e\u5f00\u5c55\u7b97\u6cd5\u7814\u7a76\u3002<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\u8981\u6c42\uff1a\u719f\u6089\u52a8\u6001\u89c4\u5212\u548c\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff1b\u719f\u7ec3\u4f7f\u7528Python\/Matlab\u7f16\u7a0b\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1a<\/span><\/section>\n
[1] Silver D, Huang A, Maddison C J, et al. Mastering the game of Go with deep neural networks and tree search[J]. nature, 2016, 529(7587): 484-489.<\/span><\/section>\n
[2] Peng Y, Chong E K P, Chen C H, et al. Ranking and selection as stochastic control[J]. IEEE Transactions on Automatic Control, 2018, 63(8): 2359-2373.<\/span><\/section>\n
[3] Li Y, Fu M C, Xu J. An Optimal Computing Budget Allocation Tree Policy for Monte Carlo Tree Search[J]. IEEE Transactions on Automatic Control, 2021.<\/span><\/section>\n<\/section>\n
\n

<\/span><\/section>\n
\u57fa\u4e8e\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u7684\u4f9b\u5e94\u94fe\u7ba1\u7406\u5b9e\u8df5<\/span><\/strong><\/span>
<\/section>\n

<\/span><\/strong><\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u5408\u4f5c\u6559\u5e08\uff1a\u5f6d\u4e00\u6770<\/span><\/section>\n

<\/section>\n
\u7814\u7a76\u5185\u5bb9\uff1a\u5f3a\u5316\u5b66\u4e60\u4efb\u52a1\u901a\u5e38\u4f7f\u7528\u9a6c\u5c14\u79d1\u592b\u51b3\u7b56\u8fc7\u7a0b<\/span>\uff08Markov Decision Process\uff0c\u7b80\u79f0MDP\uff09<\/span>\u6765\u63cf\u8ff0\uff0c\u5176\u5305\u542b\u4e24\u79cd\u6700\u57fa\u672c\u7684\u5143\u7d20\uff1a\u72b6\u6001\u4e0e\u52a8\u4f5c\uff0c\u800c\u5728\u67d0\u4e2a\u72b6\u6001\u4e0b\u6267\u884c\u67d0\u79cd\u52a8\u4f5c\uff0c\u8fd9\u4fbf\u662f\u4e00\u79cd\u7b56\u7565\u3002\u5e93\u5b58\u7ba1\u7406\u548c\u7269\u6d41\u4ea4\u901a\u7ba1\u7406\u662f\u4f9b\u5e94\u94fe\u7ba1\u7406\u4e2d\u81f3\u5173\u91cd\u8981\u7684\u4e24\u4e2a\u90e8\u5206\uff0c\u5728\u5e93\u5b58\u7ba1\u7406\u4e2d\uff0c\u56e0\u4e3a\u5e93\u5b58\u91cf\u5927\uff0c\u5e93\u5b58\u9700\u6c42\u6ce2\u52a8\u8f83\u5927\uff0c\u5e93\u5b58\u8865\u8d27\u901f\u5ea6\u7f13\u6162\u7b49\u95ee\u9898\u4f7f\u5f97\u5bf9\u5176\u7ba1\u7406\u5b58\u5728\u4e25\u5cfb\u6311\u6218\uff0c\u73b0\u6709\u7814\u7a76\u8868\u793a\uff1a\u901a\u8fc7\u7075\u6d3b\u8bbe\u5b9a\u72b6\u6001\u3001\u52a8\u4f5c\u7b49\u8981\u7d20\uff0c\u57fa\u4e8e\u5408\u7406\u7684\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u9009\u53d6\u6700\u4f18\u7b56\u7565\uff0c\u53ef\u4ee5\u8fbe\u5230\u51cf\u5c11\u5e93\u5b58\u5468\u8f6c\u65f6\u95f4\uff0c\u964d\u4f4e\u5e93\u5b58\u6210\u672c\u7b49\u76ee\u7684[1]\u3002\u7269\u6d41\u4ea4\u901a\u7ba1\u7406\u76f4\u89c2\u4e0a\u662f\u4e00\u4e2a\u52a8\u6001\u8fde\u7eed\u7684\u8fc7\u7a0b\uff0c\u4f46\u901a\u8fc7\u5408\u9002\u7684\u79bb\u6563\u5316\u5904\u7406\u4ee5\u53ca\u51fd\u6570\u903c\u8fd1\u65b9\u6cd5\u4e5f\u80fd\u591f\u4f7f\u5176\u9002\u5e94\u5f3a\u5316\u5b66\u4e60\u7684\u6846\u67b6\u3002\u672c\u8bfe\u9898\u62df\u4ece\u4e0a\u8ff0\u4e24\u65b9\u9762\u51fa\u53d1\uff0c\u7814\u7a76\u8bbe\u8ba1\u51fa\u5728\u5927\u89c4\u6a21\u72b6\u6001\u7a7a\u95f4\u4e2d\uff0c\u6bd4\u73b0\u5b58\u7b97\u6cd5\u66f4\u4f18\u7684\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff0c\u5e76\u63d0\u70bc\u51fa\u7edf\u4e00\u6027\u7684\u6846\u67b6\uff0c\u4f7f\u5176\u6210\u4e3a\u89e3\u51b3\u8be5\u7c7b\u95ee\u9898\u7684\u901a\u7528\u65b9\u6cd5\uff0c\u4ece\u800c\u63d0\u9ad8\u95ee\u9898\u89e3\u51b3\u80fd\u529b\uff0c\u51cf\u5c11\u4e3a\u7c7b\u4f3c\u5927\u89c4\u6a21\u95ee\u9898\u9010\u4e00\u7cbe\u786e\u8bbe\u8ba1\u542f\u53d1\u5f0f\u7b97\u6cd5\u7684\u65f6\u95f4\u3002<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\u8981\u6c42\uff1a\u719f\u6089\u8fd0\u7b79\u5b66\u3001\u52a8\u6001\u89c4\u5212\u548c\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff1b\u719f\u7ec3\u4f7f\u7528Python\/Matlab\u7f16\u7a0b\u3002<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e<\/span><\/section>\n
[1] Gijsbrechts, Joren, et al. “Can deep reinforcement learning improve inventory management? performance on dual sourcing, lost sales and multi-echelon problems.” Manufacturing & Service Operations Management (2021).<\/span><\/section>\n
[2] Boute, Robert N., et al. “Deep reinforcement learning for inventory control: a roadmap.” European Journal of Operational Research (2021).<\/span><\/section>\n

<\/section>\n<\/section>\n
\n
\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u8d37\u6b3e\u4fe1\u7528\u98ce\u9669\u8bc4\u4ef7<\/span><\/strong><\/span>
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/span><\/section>\n
\u5408\u4f5c\u6559\u5e08\uff1a\u5f6d\u4e00\u6770<\/span><\/section>\n

<\/section>\n
\u7814\u7a76\u5185\u5bb9\uff1a\u91d1\u878d\u4f5c\u4e3a\u73b0\u4ee3\u7ecf\u6d4e\u7684\u6838\u5fc3\uff0c\u5728\u4e61\u6751\u632f\u5174\u591a\u5143\u6295\u5165\u683c\u5c40\u4e2d\u5177\u6709\u57fa\u7840\u5730\u4f4d\u548c\u652f\u6491\u6760\u6746\u4f5c\u7528\uff0c\u5176\u98ce\u9669\u6cbb\u7406\u6c34\u5e73\u76f4\u63a5\u5f71\u54cd\u7740\u65b0\u65f6\u671f\u519c\u4e1a\u519c\u6751\u9ad8\u8d28\u91cf\u53d1\u5c55\u7684\u7a33\u5b9a\u6027\u3002\u5341\u516b\u5927\u4ee5\u6765\uff0c\u4e2d\u592e\u653f\u5e9c\u591a\u6b21\u5f3a\u8c03\u9632\u8303\u5316\u89e3\u91cd\u5927\u98ce\u9669\uff0c\u5c24\u5176\u662f\u91d1\u878d\u98ce\u9669\uff0c\u5e76\u76f8\u7ee7\u9881\u5e03\u4e86\u7cfb\u5217\u6cbb\u7406\u63aa\u65bd\u3002\u8fd1\u4e9b\u5e74\uff0c\u519c\u6751\u91d1\u878d\u4ea7\u54c1\u4e0e\u670d\u52a1\u7684\u521b\u65b0\u6b65\u4f10\u4e0d\u65ad\u52a0\u5feb\uff0c\u519c\u6751\u91d1\u878d\u6539\u9769\u521b\u65b0\u8fdb\u7a0b\u4e2d\u7684\u6f5c\u5728\u98ce\u9669\u95ee\u9898\u503c\u5f97\u9ad8\u5ea6\u91cd\u89c6\u3002\u4f5c\u4e3a\u519c\u5730\u201c\u4e09\u6743\u5206\u7f6e\u201d\u5236\u5ea6\u6846\u67b6\u4e0b\u6211\u56fd\u519c\u6751\u91d1\u878d\u6539\u9769\u7684\u91cd\u8981\u521b\u65b0\uff0c\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u878d\u8d44\u5728\u89e3\u51b3\u519c\u6237\u8d37\u6b3e\u62b5\u62bc\u96be\u3001\u62c5\u4fdd\u96be\u95ee\u9898\uff0c\u76d8\u6d3b\u519c\u6751\u5b58\u91cf\u8d44\u4ea7\u3001\u6d3b\u8dc3\u519c\u6751\u7ecf\u6d4e\u65b9\u9762\u53d1\u6325\u4e86\u663e\u8457\u4f5c\u7528\u3002\u4e0e\u6b64\u540c\u65f6\uff0c\u4e0d\u5bb9\u5ffd\u89c6\u7684\u662f\uff0c\u56e0\u53d7\u9650\u4e8e\u571f\u5730\u4ea7\u6743\u4ea4\u6613\u5e02\u573a\u53d1\u5c55\u6ede\u540e\u3001\u519c\u4e1a\u7ecf\u8425\u4e3b\u4f53\u4fe1\u7528\u4f53\u7cfb\u4e0d\u5b8c\u5584\u3001\u62b5\u62bc\u7269\u4ef7\u503c\u4e0d\u7a33\u5b9a\u4ee5\u53ca\u98ce\u9669\u7ba1\u7406\u4f53\u7cfb\u4e0d\u5065\u5168\u7b49\u591a\u91cd\u56e0\u7d20\uff0c\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u8d37\u6b3e\u4e1a\u52a1\u5f00\u5c55\u9762\u4e34\u7740\u98ce\u9669\u4e0e\u6536\u76ca\u4e0d\u5bf9\u7b49\u7684\u56f0\u5883\u3002\u524d\u671f\u8bd5\u70b9\u5730\u533a\u7684\u90e8\u5206\u94f6\u884c\u6216\u4fe1\u7528\u793e\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u8d37\u6b3e\u4f9b\u7ed9\u5b58\u5728\u89c4\u6a21\u6536\u7f29\u6216\u57fa\u672c\u505c\u6ede\u73b0\u8c61\u3002\u5728\u6b64\u80cc\u666f\u4e0b\uff0c\u7cfb\u7edf\u5f00\u5c55\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u878d\u8d44\u98ce\u9669\u8bc4\u4f30\uff0c\u5c24\u5176\u662f\u519c\u5730\u62b5\u62bc\u8d37\u6b3e\u4e1a\u52a1\u7684\u6838\u5fc3\u98ce\u9669\u2014\u2014\u4fe1\u7528\u98ce\u9669\u8bc4\u4ef7\u7814\u7a76\uff0c\u5bf9\u4e8e\u7834\u89e3\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u8d37\u6b3e\u75c7\u7ed3\u3001\u63d0\u9ad8\u519c\u5730\u91d1\u878d\u5e02\u573a\u6709\u6548\u6027\u5177\u6709\u91cd\u8981\u73b0\u5b9e\u610f\u4e49\u3002\u68b3\u7406\u5df2\u6709\u7814\u7a76\u770b\uff0c\u5df2\u6709\u7814\u7a76\u96c6\u4e2d\u4e8e\u519c\u5730\u62b5\u62bc\u8d37\u6b3e\u98ce\u9669\u5206\u7c7b\u3001\u5f15\u81f4\u56e0\u7d20\u53ca\u9632\u63a7\u63aa\u65bd\u7684\u5b9a\u6027\u5206\u6790\uff0c\u4ec5\u6709\u5c11\u91cf\u5b66\u8005\u5bf9\u519c\u5730\u62b5\u62bc\u8d37\u6b3e\u7684\u98ce\u9669\u56e0\u7d20\u8fdb\u884c\u4e86\u8bc6\u522b\u548c\u8bc4\u4ef7\uff0c\u53e6\u6709\u4e2a\u522b\u7814\u7a76\u8fd0\u7528Logistic\u56de\u5f52\u5206\u6790\u519c\u5730\u62b5\u62bc\u8d37\u6b3e\u4fe1\u7528\u98ce\u9669\u5f71\u54cd\u56e0\u7d20\u5e76\u9884\u6d4b\u8fdd\u7ea6\u6982\u7387\uff0c\u540c\u65f6\u4f9d\u636eCreditRisk+\u6a21\u578b\uff0c\u5bf9\u4fe1\u7528\u98ce\u9669\u8861\u91cf\u8fdb\u884c\u4e86\u7814\u7a76\u3002\u7136\u800c\uff0c\u5df2\u6709\u7814\u7a76\u5b58\u5728\u4e24\u65b9\u9762\u7684\u4e0d\u8db3\uff1a\u4e00\u662f\u519c\u5730\u62b5\u62bc\u8d37\u6b3e\u4fe1\u7528\u98ce\u9669\u7684\u5173\u952e\u53c2\u6570\u4f30\u8ba1\u65b9\u6cd5\u5b58\u5728\u4e0d\u8db3\uff0c\u8fd0\u7528Logistic\u56de\u5f52\u9884\u6d4b\u8fdd\u7ea6\u6982\u7387\u5f80\u5f80\u5b58\u5728\u201c\u4fe1\u7528\u5f97\u5206\u5f88\u9ad8\uff0c\u4f46\u8fdd\u7ea6\u7387\u8f83\u9ad8\u201d\u7684\u6096\u8bba\uff1b\u4e8c\u662f\u73b0\u6709\u7684\u4e2a\u4f53\u4fe1\u7528\u98ce\u9669\u7814\u7a76\u5927\u591a\u53ea\u662f\u5f97\u51fa\u4e2a\u4f53\u4fe1\u7528\u8bc4\u5206\uff0c\u5c1a\u4e0d\u80fd\u5b9e\u73b0\u5728\u4f30\u7b97\u51fa\u91d1\u878d\u6d88\u8d39\u8005\u7684\u4fe1\u7528\u5f97\u5206\u57fa\u7840\u4e0a\uff0c\u8fdb\u4e00\u6b65\u8ba1\u7b97\u51fa\u8d37\u4e0e\u4e0d\u8d37\u7684\u4e34\u754c\u503c\u3002\u800c\u8fd9\u4e2a\u4e34\u754c\u503c\u624d\u662f\u771f\u6b63\u53ef\u4ee5\u4e3a\u91d1\u878d\u673a\u6784\u63d0\u4f9b\u8d37\u6b3e\u51b3\u7b56\u53c2\u8003\u7684\u4f9d\u636e\u3002<\/span><\/section>\n

<\/section>\n
\u7814\u7a76\u76ee\u6807\uff1a<\/span><\/section>\n
\uff081\uff09\u5229\u7528\u673a\u5668\u5b66\u4e60\u76f8\u5173\u65b9\u6cd5\uff08\u5982\u968f\u673a\u68ee\u6797\u3001\u6df1\u5ea6\u68ee\u6797\u7b49\uff09\u4f18\u5316\u4fe1\u7528\u98ce\u9669\u7684\u5173\u952e\u53c2\u6570\u4f30\u8ba1\uff0c\u4ee5\u63d0\u9ad8\u4fe1\u7528\u5f97\u5206\u8bc4\u4ef7\u6709\u6548\u6027\uff1b<\/span><\/section>\n
\uff082\uff09\u53ef\u4ee5\u5728\u4fe1\u7528\u8bc4\u5206\u57fa\u7840\u4e0a\uff0c\u4f30\u7b97\u51fa\u8d37\u4e0e\u4e0d\u8d37\u7684\u4e34\u754c\u503c\uff08\u53ef\u4ee5\u7528\u7cfb\u7edf\u4eff\u771f\u8fdb\u884c\u6a21\u62df\uff09\u3002<\/span><\/section>\n

<\/section>\n
\u8bfe\u9898\u8981\u6c42\uff1a<\/span><\/section>\n
\uff081\uff09\u5177\u6709\u91d1\u878d\u98ce\u9669\u76f8\u5173\u7684\u7814\u7a76\u57fa\u7840 <\/span><\/section>\n
\uff082\uff09\u4f1a\u8fd0\u7528\u673a\u5668\u5b66\u4e60\u3001\u7cfb\u7edf\u4eff\u771f\u76f8\u5173\u65b9\u6cd5<\/span><\/section>\n

<\/section>\n
\u53c2\u8003\u6587\u732e\uff1a<\/span><\/section>\n
[1]\u5415\u5fb7\u5b8f\uff0c\u5f20\u65e0\u5777.\u519c\u5730\u7ecf\u8425\u6743\u62b5\u62bc\u8d37\u6b3e\u4fe1\u7528\u98ce\u9669\u5f71\u54cd\u56e0\u7d20\u53ca\u5176\u8861\u91cf\u7814\u7a76\u2014\u2014\u57fa\u4e8eCreditRisk+\u6a21\u578b\u7684\u4f30\u8ba1\uff0c\u534e\u4e2d\u519c\u4e1a\u5927\u5b66\u5b66\u62a5\uff08\u793e\u4f1a\u79d1\u5b66\u7248\uff09\uff0c2018\uff084\uff09\uff1a137-147.<\/span><\/section>\n
[2] Milad Malekipirbazari, Vural Aksakalli. Risk assessment in social lending via random forests. Expert Systems with Application,2015,42:4621-4631.<\/span><\/section>\n
[3] X. Ma, J. Sha, D. Wang, Y. Yu, Q. Yang, X. Niu, Study on A Prediction of P2P Network Loan Default Based on the Machine Learning LightGBM and XGboost Algorithms according to Different High Dimensional Data Cleaning, Electronic Commerce Research and Applications (2018), doi: https:\/\/doi.org\/10.1016\/j.elerap.2018.08.002<\/span><\/section>\n

<\/span><\/section>\n

<\/span><\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

\u7533\u8bf7\u65b9\u5f0f\u53ca\u5176\u4ed6<\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n
\n
\u7533\u8bf7\u65b9\u5f0f<\/span><\/strong><\/span><\/section>\n

<\/span><\/section>\n
\u8bf7\u5c06\u7b80\u5386\u53d1\u9001\u81f3\uff1acmar@pku.edu.cn<\/span><\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n
\u90ae\u4ef6\u6807\u9898\uff1aAI\u9662\u591a\u667a\u80fd\u4f53\u4e2d\u5fc3\u5b66\u672f\u51ac\u4ee4\u8425+\u9879\u76ee\u540d\u79f0+\u5408\u4f5c\u6559\u5e08\u59d3\u540d<\/span><\/section>\n<\/section>\n
\n

<\/section>\n<\/section>\n
\n
\u4e00\u5219\u62db\u8058<\/span><\/strong><\/span><\/section>\n

<\/span><\/section>\n
\u4ec0\u4e48\uff1f\u770b\u4e86\u4ee5\u4e0a\u7684\u9879\u76ee\u4fe1\u606f\uff0c\u4f60\u4e0d\u4ec5\u60f3\u53c2\u52a0\u4e2d\u5fc3\u7684\u9879\u76ee\uff0c\u8fd8\u60f3\u52a0\u5165\u5317\u5927\u4e0e\u4f17\u591a\u8001\u5e08\u4eec\u4e00\u8d77\u4ece\u4e8b\u7814\u7a76\uff1f<\/span><\/section>\n<\/section>\n
\n

<\/section>\n
\u5149\u534e\u601d\u60f3\u529b\u7814\u7a76\u5458\u62db\u52df\u5566\uff5e<\/span><\/section>\n

<\/section>\n
\u5149\u534e\u601d\u60f3\u529b\u7814\u7a76\u5458\u5c5e\u4e8e\u5149\u534e\u7ba1\u7406\u5b66\u9662\u9662\u8058\u5408\u540c\u5236\u5458\u5de5\uff0c\u53ef\u77ed\u671f\uff08\u4e00\u5e74\uff09\u6216\u957f\u671f\u4ece\u4e8b\u7814\u7a76\u3002\u8981\u6c42\u4e3a\u8ba1\u7b97\u673a\u76f8\u5173\u4e13\u4e1a\u7684\u7814\u7a76\u751f\uff0c\u85aa\u916c\u548c\u798f\u5229\u9762\u8bae\uff0c\u6709\u610f\u8005\u8bf7\u8054\u7cfbcmar@pku.edu.cn\u3002<\/span><\/section>\n<\/section>\n
\n

<\/span><\/section>\n
\u671f\u5f85\u4f60\u7684\u52a0\u5165\uff5e<\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section><\/div>\n","protected":false},"excerpt":{"rendered":"

\u5bfc\u8bed \u5317\u4eac\u5927\u5b66\u4eba\u5de5\u667a\u80fd\u7814\u7a76\u9662\u591a\u667a\u80fd\u4f53\u4e2d\u5fc3\u5b66\u672f\u51ac\u4ee4\u8425\u5f00\u59cb\u62db\u52df\u5566\uff5e \u5728\u8fd9\u91cc\uff0c\u4f60\u4e0d\u4ec5\u53ef\u4ee5\u8ddf\u968f\u4e2d\u5fc3\u7684\u8001\u5e08\u4eec\u8fdb\u884c\u9879\u76ee\u7684\u7814\u7a76\u4e0e\u5b66\u4e60\uff0c\u589e\u52a0\u4f60\u7684\u79d1\u7814\u5b9e\u8df5\u7ecf\u5386\uff1b\u66f4\u80fd\u4e30\u5bcc\u4eba\u751f\u4f53\u9a8c\uff0c\u63d0\u9ad8\u4f60\u7684\u72ec\u7acb\u81ea\u4e3b\u80fd\u529b\u548c\u521b\u65b0\u7cbe\u795e\u3002 2022~\u5c31\u8ba9\u5317\u4eac\u5927\u5b66\u4eba\u5de5\u667a\u80fd\u7814\u7a76\u9662\u591a\u667a\u80fd\u4f53\u4e2d\u5fc3\u5b66\u672f\u51ac\u4ee4\u8425\u966a\u4f34\u4f60\u5f00\u542f\u65b0\u7684\u4e00\u5e74\u5427\uff5e \u5317\u4eac\u5927\u5b66\u4eba\u5de5\u667a\u80fd\u7814\u7a76\u9662 …<\/p>\n","protected":false},"author":1,"featured_media":32059,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"special":[],"_links":{"self":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/32061"}],"collection":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=32061"}],"version-history":[{"count":0,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/32061\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/media\/32059"}],"wp:attachment":[{"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=32061"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=32061"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=32061"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fspecial&post=32061"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}