{"id":59921,"date":"2025-05-21T19:42:28","date_gmt":"2025-05-21T11:42:28","guid":{"rendered":"https:\/\/swarma.org\/?p=59921"},"modified":"2025-05-21T19:42:28","modified_gmt":"2025-05-21T11:42:28","slug":"%e5%86%b3%e7%ad%96%e4%b9%8b%e6%a1%a5%ef%bc%9a%e6%95%b0%e6%8d%ae%e9%a9%b1%e5%8a%a8%e4%bc%98%e5%8c%96%e6%96%b9%e6%b3%95%e7%9a%84%e8%bf%9b%e5%b1%95%e4%b8%8e%e6%8c%91%e6%88%98","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=59921","title":{"rendered":"\u51b3\u7b56\u4e4b\u6865\uff1a\u6570\u636e\u9a71\u52a8\u4f18\u5316\u65b9\u6cd5\u7684\u8fdb\u5c55\u4e0e\u6311\u6218"},"content":{"rendered":"
\n
\n
\n
\n
\n
\n
\n
<\/section>\n
\n
\n
\n
\n
\n

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

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

\n

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

\u5728\u5f53\u4eca\u5927\u6570\u636e\u65f6\u4ee3\uff0c\u6570\u636e\u5df2\u6210\u4e3a\u63a8\u52a8\u51b3\u7b56\u7684\u6838\u5fc3\u8d44\u6e90\u3002\u7136\u800c\uff0c\u5982\u4f55\u5c06\u6d77\u91cf\u6570\u636e\u8f6c\u5316\u4e3a\u6700\u4f18\u51b3\u7b56\uff0c\u4ecd\u662f\u5b66\u672f\u754c\u548c\u5de5\u4e1a\u754c\u9762\u4e34\u7684\u91cd\u5927\u6311\u6218\u3002\u8fd1\u65e5\u5317\u4eac\u5927\u5b66\u5b8b\u6d01\u6559\u6388\u56e2\u961f\u5728Cell Press\u4ea4\u53c9\u5b66\u79d1\u671f\u520aNexus\u53d1\u8868\u9898\u4e3a\u300aBridging Prediction and Decision: Advances and Challenges in Data-Driven Optimization\u300b\u7684\u89c2\u70b9\u6027\u7efc\u8ff0\u6587\u7ae0\uff0c\u7cfb\u7edf\u63a2\u8ba8\u4e86\u6570\u636e\u9a71\u52a8\u7684\u65b9\u6cd5\u5728\u9884\u6d4b\u548c\u51b3\u7b56\u9886\u57df\u7684\u5e94\u7528\u3002\u6587\u7ae0\u7cfb\u7edf\u603b\u7ed3\u4e86\u5e8f\u8d2f\u4f18\u5316\uff08SO\uff09\u3001\u7aef\u5230\u7aef\u5b66\u4e60\uff08E2E\uff09\u548c\u76f4\u63a5\u5b66\u4e60\uff08DL\uff09\u4e09\u79cd\u6570\u636e\u9a71\u52a8\u578b\u4f18\u5316\u65b9\u6cd5\u7684\u7406\u8bba\u3001\u4f18\u52bf\u548c\u6700\u65b0\u8fdb\u5c55\uff0c\u5e76\u7ed3\u5408\u80fd\u6e90\u8c03\u5ea6\u3001\u8fd0\u7b79\u7ba1\u7406\u548c\u667a\u80fd\u63a7\u5236\u7b49\u5178\u578b\u5e94\u7528\u573a\u666f\u8fdb\u884c\u4e86\u5206\u6790\uff0c\u901a\u8fc7\u5256\u6790\u6570\u636e\u8d28\u91cf\u3001\u4f18\u5316\u5efa\u6a21\u548c\u51b3\u7b56\u5e94\u7528\u7b49\u5173\u952e\u95ee\u9898\uff0c\u6587\u7ae0\u63d0\u51fa\u4e86\u5bf9\u5e94\u7684\u89e3\u51b3\u65b9\u6848\uff0c\u4e3a\u9762\u5411\u590d\u6742\u73af\u5883\u7684\u667a\u80fd\u51b3\u7b56\u63d0\u4f9b\u4e86\u91cd\u8981\u7684\u7406\u8bba\u652f\u6491\u548c\u65b9\u6cd5\u6307\u5f15\u3002<\/span><\/span><\/strong><\/p>\n


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

\u5173\u952e\u8bcd\uff1a\u4f18\u5316\uff0c\u5e8f\u8d2f\u4f18\u5316\uff08Sequential Optimization\uff0cSO\uff09\uff0c\u7aef\u5230\u7aef\u5b66\u4e60\uff08End to End Learning, E2E\uff09\uff0c\u76f4\u63a5\u5b66\u4e60\uff08Direct Learning, DL\uff09<\/span>
<\/span><\/span><\/strong><\/p>\n<\/section>\n

<\/strong><\/strong><\/section>\n<\/section>\n
\n
<\/strong><\/strong><\/section>\n<\/section>\n
\n
<\/strong><\/strong><\/section>\n
<\/strong><\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\"\u56fe\u7247\"<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\u9646\u6021\u821f\u4e28\u4f5c\u8005<\/span><\/span><\/p>\n

\u66fe\u5229\u4e28\u5ba1\u6821<\/span><\/span><\/p>\n


<\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n
\n

\u8bba\u6587\u9898\u76ee\uff1aBridging prediction and decision: Advances and challenges in data-driven optimization <\/span><\/p>\n

\u8bba\u6587\u5730\u5740\uff1ahttps:\/\/www.cell.com\/nexus\/fulltext\/S2950-1601(25)00004-X<\/span><\/p>\n

\u8bba\u6587\u6765\u6e90\uff1aNexus<\/span><\/p>\n<\/blockquote>\n


<\/span><\/p>\n


<\/span><\/p>\n

\n
\n
\n
\n
\n

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

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

\u80cc\u666f\u77e5\u8bc6<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

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

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


<\/span><\/p>\n

\u4ece\u77f3\u5668\u65f6\u4ee3\u7684\u751f\u5b58\u9009\u62e9\u5230\u6570\u5b57\u65f6\u4ee3\u7684\u667a\u80fd\u51b3\u7b56\uff0c\u4eba\u7c7b\u6587\u660e\u7684\u53d1\u5c55\u53f2\u4e5f\u662f\u4e00\u90e8\u51b3\u7b56\u80fd\u529b\u7684\u8fdb\u5316\u53f2\u3002\u968f\u7740\u51b3\u7b56\u79d1\u5b66\u6b63<\/span><\/span>\u7ecf\u5386\u4ece”\u7ecf\u9a8c\u9a71\u52a8”\u5230”\u6570\u636e\u9a71\u52a8”\u7684\u8303\u5f0f\u9769\u547d\uff0c\u73b0\u4ee3\u51b3\u7b56\u95ee\u9898\u8d8a\u6765\u8d8a\u4f9d\u8d56\u4e8e\u7cfb\u7edf\u5316\u7684\u89e3\u51b3\u65b9\u6cd5<\/span><\/span><\/strong>\u2014\u2014\u901a\u8fc7\u5c06\u590d\u6742\u7684\u51b3\u7b56\u9700\u6c42\u8f6c\u5316\u4e3a\u53ef\u8ba1\u7b97\u7684\u4f18\u5316\u95ee\u9898\uff0c\u6211\u4eec\u5f97\u4ee5\u5728\u6570\u636e\u6d77\u6d0b\u4e2d\u5bfb\u627e\u6700\u4f18\u89e3\uff0c\u4e0b\u9762\u8ba9\u6211\u4eec\u901a\u8fc7\u4e24\u4e2a\u5178\u578b\u51b3\u7b56\u95ee\u9898\u6765\u6df1\u5165\u7406\u89e3\u8fd9\u4e00\u8fc7\u7a0b\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u88c5\u8f7d\u95ee\u9898\uff08\u4e2a\u4eba\u51b3\u7b56\uff09<\/span><\/span><\/strong>\uff1a\u5047\u8bbe\u6709\u4e00\u8f86\u8f7d\u91cd\u91cf\u4e0a\u9650\u4e3aB<\/span><\/span>\u7684\u8d27\u8f66\u6765\u5230\u8d27\u573a\uff0c\u9700\u8981\u4ecen<\/span><\/span><\/span><\/span>\u4ef6\u8d27\u54c1\u4e2d\u9009\u62e9\u90e8\u5206\u8d27\u7269\u8fdb\u884c\u88c5\u8f7d\u3002\u6bcf\u4ef6\u8d27\u7269<\/span><\/span><\/span><\/span>i<\/span><\/span><\/span><\/span>\u7684\u91cd\u91cf\u4e3a<\/span><\/span>a<\/span><\/span>i<\/span><\/sub><\/span><\/span><\/span>\uff0c\u4ef7\u503c\u4e3a<\/span><\/span>c<\/span><\/span>i<\/span><\/sub><\/span><\/span><\/span>\u3002\u8f66\u4e3b\u7684\u76ee\u6807\u662f\u5728\u4e0d\u8d85\u8fc7\u8d27\u8f66\u6700\u5927\u8f7d\u91cd\u91cf<\/span><\/span>B<\/span><\/span>\u7684\u524d\u63d0\u4e0b\uff0c\u9009\u62e9\u82e5\u5e72\u4ef6\u8d27\u7269\uff0c\u4f7f\u5f97\u88ab\u9009\u4e2d\u7684\u8d27\u7269\u7684\u603b\u4ef7\u503c\u6700\u5927\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u5982\u679c\u5b9a\u4e49\u8fd9\u6837\u7684\u4e00\u4e2a\u8f85\u52a9\u51b3\u7b56\u53d8\u91cf\uff1a<\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\uff0c\u90a3\u4e48\u88c5\u8f7d\u95ee\u9898\u53ef\u4ee5\u5199\u4e3a\u5982\u4e0b\u7684\u4f18\u5316\u95ee\u9898\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n


<\/span><\/span><\/span><\/p>\n

\u7535\u7f51<\/span><\/span><\/strong>\u6700\u4f18\u6f6e\u6d41<\/span><\/span><\/strong>\u63a7\u5236\u95ee\u9898\uff08\u4f01\u4e1a\u51b3\u7b56\uff09<\/span><\/span><\/strong>\uff1a\u5728\u7535\u529b\u7cfb\u7edf\u8fd0\u884c\u4e2d\uff0c\u6700\u4f18\u6f6e\u6d41\u63a7\u5236\u95ee\u9898\u7684\u6838\u5fc3\u76ee\u6807\u662f\u901a\u8fc7\u8c03\u6574\u53d1\u7535\u673a\u7ec4\u7684\u8f93\u51fa\u529f\u7387\u7b49\u53ef\u63a7\u53c2\u6570\uff0c\u5728\u6ee1\u8db3\u8282\u70b9\u6b63\u5e38\u529f\u7387\u5e73\u8861\u53ca\u5404\u79cd\u5b89\u5168\u6307\u6807\u7684\u7ea6\u675f\u4e0b\uff0c\u5b9e\u73b0\u7cfb\u7edf\u8fd0\u884c\u7684\u76ee\u6807\u51fd\u6570\u6700\u5c0f\u5316\u3002\u5047\u5982\u67d0\u706b\u7535\u5382\u6709n<\/span><\/span>\u53f0\u706b\u7535\u673a\u7ec4\uff0c\u6bcf\u53f0\u673a\u7ec4\u7684\u8f93\u51fa\u529f\u7387\u4e3az<\/span><\/span><\/span><\/span>i<\/span><\/span><\/sub><\/span><\/span>\uff0c\u673a\u7ec4\u7684\u6210\u672c\u51fd\u6570\u662f\u8f93\u51fa\u529f\u7387\u7684\u4e8c\u6b21\u51fd\u6570\uff0c\u5373<\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\u3002\u706b\u7535\u5382\u9762\u4e34\u4e24\u4e2a\u7ea6\u675f\uff0c\u4e00\u662f\u6bcf\u53f0\u673a\u7ec4\u7684\u8f93\u51fa\u529f\u7387\u9700\u8981\u4fdd\u6301\u5728\u6700\u4f73\u8fd0\u884c\u529f\u7387\u533a\u95f4<\/span><\/span>\"image.png\"<\/span><\/sub>\u4e4b\u5185\uff0c\u540c\u65f6\uff0c\u6240\u6709\u673a\u7ec4\u7684\u603b\u8f93\u51fa\u529f\u7387\u4e0d\u80fd\u4f4e\u4e8e\u67d0\u4e2a\u6570\u503cL<\/span><\/span>\u3002\u540c\u6837\u5730\uff0c\u706b\u7535\u5382\u7684\u51b3\u7b56\u95ee\u9898\u4e5f\u53ef\u4ee5\u5199\u4e3a\u4e00\u4e2a\u4f18\u5316\u95ee\u9898\u3002<\/span><\/span><\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n


<\/span><\/span><\/span><\/p>\n

\u4ece\u524d\u8ff0\u7684\u4e24\u4e2a\u51b3\u7b56\u95ee\u9898\u7684\u4f8b\u5b50\u51fa\u53d1\uff0c\u53ef\u4ee5\u5c06\u51b3\u7b56\u95ee\u9898\u4e00\u822c\u5316\u4e3a\u5982\u4e0b\u5f62\u5f0f\uff1a<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n


<\/span><\/span><\/span><\/p>\n

\u8fd9\u91cc\u7684z<\/span><\/span>\u4ee3\u8868\u4e86\u51b3\u7b56\u53d8\u91cf\uff0c<\/span><\/span><\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\u53ef\u4ee5\u8fdb\u4e00\u6b65\u53c2\u6570\u5316\u4e3a<\/span><\/span>p<\/span><\/span>\u03b8<\/span><\/sub>(x)<\/span><\/span><\/span><\/span><\/span>\uff0c\u4ee3\u8868\u4e86\u4ece\u8f93\u5165\u53d8\u91cf\uff08\u73af\u5883\u53d8\u91cf\uff09x<\/span><\/span>\u5bf9\u67d0\u4e2a\u76f8\u5173\u53d8\u91cf\uff08\u4e2d\u95f4\u53d8\u91cf\u3001\u5185\u90e8\u53d8\u91cf\uff09y<\/span><\/span><\/span><\/span>\u7684\u9884\u6d4b\u503c\u3002\u5e76\u4e14\u6ce8\u610f\u5230\uff0c\u8fd9\u4e2a\u9884\u6d4b\u503c<\/span><\/span><\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\u53ef\u80fd\u4e0e\u771f\u5b9e\u503cy<\/span><\/span><\/span><\/span>\u76f8\u7b49\uff0c\u4e5f\u53ef\u80fd\u4e0e\u771f\u5b9e\u503c<\/span><\/span><\/span><\/span>y<\/span><\/span><\/span><\/span>\u4e4b\u95f4\u5b58\u5728\u504f\u5dee\u3002\u5bf9\u88c5\u8f7d\u95ee\u9898\u6765\u8bf4\uff0c\u4e0d\u540c\u8d27\u7269\u7684\u91cd\u91cf\u548c\u4ef7\u503c\u5747\u4e8b\u5148\u5df2\u77e5\uff0c<\/span><\/span>\"image.png\"<\/span><\/sub>\uff0c\u9884\u6d4b\u503c\u4e0e\u771f\u5b9e\u503c\u662f\u76f8\u7b49\u7684\u3002\u5bf9\u4e8e\u7535\u7f51\u63a7\u5236\u95ee\u9898\u6765\u8bf4\uff0c\u6210\u672c\u65b9\u7a0b\uff083\uff09\u7684\u5404\u9879\u7cfb\u6570\u662f\u672a\u77e5\u7684\uff0c\u9884\u6d4b<\/span><\/span><\/span><\/span>\u503c<\/span><\/span><\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\u4e0e\u771f\u5b9e\u503c<\/span><\/span><\/span><\/span>y={a<\/span><\/span>i<\/span><\/sub>,b<\/span><\/span>i<\/span><\/sub>,c<\/span><\/span><\/span><\/span>i<\/span><\/sub>},i=1, \u2026,n<\/span><\/span><\/span><\/span><\/span> \u5f88\u53ef\u80fd\u4e0d\u76f8\u7b49\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u5728\u590d\u6742\u7684\u73b0\u5b9e\u51b3\u7b56\u95ee\u9898\u4e2d\uff0c\u6240\u9700\u8981\u9884\u6d4b\u7684\u53d8\u91cfy<\/span><\/span>\u7684\u5f62\u5f0f\u66f4\u52a0\u591a\u6837\uff0c\u9884\u6d4b\u95ee\u9898\u66f4\u52a0\u56f0\u96be\u3002\u4f8b\u5982\uff0c\u5728\u80fd\u6e90\u5e02\u573a\u7684\u76f8\u5173\u51b3\u7b56\u95ee\u9898\u4e2d\uff0c\u901a\u5e38\u9700\u8981\u5bf9\u7535\u529b\u4ef7\u683c\u3001\u7535\u529b\u9700\u6c42\u3001\u98ce\u5149\u65b0\u80fd\u6e90\u53d1\u7535\u91cf\u7b49\u8fdb\u884c\u9884\u6d4b\uff0c\u5f80\u5f80\u4f1a\u6d89\u53ca\u590d\u6742\u7684\u7ecf\u6d4e\u793e\u4f1a\u7cfb\u7edf\u3001\u6c14\u8c61\u7cfb\u7edf\u7684\u9884\u6d4b\u95ee\u9898\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u968f\u7740\u673a\u5668\u5b66\u4e60\u548c\u6df1\u5ea6\u5b66\u4e60\u6280\u672f\u7684\u98de\u901f\u53d1\u5c55\uff0c\u201c\u9884\u6d4b+\u51b3\u7b56\u201d\u6846\u67b6\u4f5c\u4e3a\u4eba\u5de5\u667a\u80fd\u4e0e\u7cfb\u7edf\u79d1\u5b66\u76f8\u7ed3\u5408\u7684\u91cd\u8981\u624b\u6bb5\uff0c\u9010\u6e10\u6210\u4e3a\u63d0\u5347\u7cfb\u7edf\u6548\u80fd\u7684\u6838\u5fc3\u6311\u6218\uff0c\u4e5f\u662f\u6700\u7ec8\u5b9e\u73b0\u5927\u6570\u636e\u4ef7\u503c\u8f6c\u5316\u7684\u5173\u952e\u9014\u5f84\u3002\u5b8b\u6d01\u6559\u6388\u56e2\u961f\u7684\u7efc\u8ff0\u6587\u7ae0\u4ece\u7406\u8bba\u3001\u5e94\u7528\u3001\u6311\u6218\u548c\u89e3\u51b3\u65b9\u6848\u56db\u4e2a\u7ef4\u5ea6\u6df1\u5165\u5206\u6790\u6570\u636e\u9a71\u52a8\u7684\u4f18\u5316\u65b9\u6cd5\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n


<\/span><\/h2>\n

\n
\n
\n
\n
\n

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

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

\u4ece\u9884\u6d4b\u5230\u51b3\u7b56\uff1a\u6570\u636e\u9a71\u52a8\u4f18\u5316\u7684\u4e09\u5927\u8303\u5f0f<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

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

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


<\/span><\/p>\n

<\/section>\n

\u56fe1\uff1a\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\uff08\u8303\u5f0f\u4e00\uff09<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\uff08Sequential Optimization\uff0cSO\uff09<\/span><\/span><\/strong>\uff1a<\/span><\/span><\/strong>\u8be5\u65b9\u6cd5\u91c7\u7528\u201c\u9884\u6d4b\u201d-\u201c\u4f18\u5316\u201d\u4e24\u9636\u6bb5\u6d41\u7a0b\u6765\u89e3\u51b3\u51b3\u7b56\u95ee\u9898\uff1a\u9996\u5148\u5229\u7528\u9884\u6d4b\u6a21\u578b\u5bf9\u4e0d\u786e\u5b9a\u56e0\u7d20\u8fdb\u884c\u4f30\u8ba1\uff0c\u518d\u5c06\u9884\u6d4b\u7ed3\u679c\u8f93\u5165\u4f18\u5316\u6a21\u5757\u751f\u6210\u51b3\u7b56\u3002\u6b64\u65b9\u6cd5\u76f4\u89c2\u4e14\u6a21\u5757\u5316\uff0c\u5404\u73af\u8282\u53ef\u72ec\u7acb\u4f18\u5316\uff0c\u5b9e\u73b0\u4eba\u5de5\u667a\u80fd\u4e0e\u51b3\u7b56\u6a21\u578b\u7684\u76f4\u63a5\u8026\u5408\u3002\u5e38\u7528\u7684\u5b9e\u73b0\u4e3a\u7b2c\u4e00\u6b65\u662f\u901a\u8fc7\u6570\u636e\u9a71\u52a8\u7684\u65b9\u6cd5\uff0c\u53d1\u73b0\u6570\u636e\u6240\u4ee3\u8868\u7684\u590d\u6742\u7cfb\u7edf\u7684\u6f5c\u5728\u8fd0\u884c\u6a21\u5f0f\uff0c\u4ece\u800c\u5f97\u5230\u9884\u6d4b\u503c<\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\u3002\u7b2c\u4e8c\u6b65\u518d\u501f\u52a9Gurobi\u7b49\u4f18\u5316\u6c42\u89e3\u5668\uff0c\u6c42\u89e3\u4f18\u5316\uff08\u516c\u5f0f 6-7\uff09\u3002<\/span><\/span><\/span><\/span><\/span><\/span><\/p>\n


<\/span><\/p>\n

\u4f46\u662f\uff0c\u7531\u4e8e\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u5c06\u51b3\u7b56\u95ee\u9898\u89e3\u8026\u6210\u4e24\u4e2a\u5b50\u95ee\u9898\uff0c\u5f88\u53ef\u80fd\u4ea7\u751f\u4f18\u5316\u9636\u6bb5\u548c\u51b3\u7b56\u9636\u6bb5\u4e4b\u95f4\u7684\u4e0d\u4e00\u81f4\u60c5\u51b5\u3002\u5728\u4f18\u5316\u9636\u6bb5\uff0c\u7531\u4e8e\u9884\u6d4b\u7684\u7cbe\u5ea6\u6709\u9650\uff0c\u9884\u6d4b<\/span><\/span>\u503c<\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span><\/span>\u5f80\u5f80\u662f\u5bf9\u79f0\u5206\u5e03\u5728\u771f\u5b9e\u503cy<\/span><\/span>\u7684\u5468\u56f4\u3002\u4f46\u662f\u51b3\u7b56\u9636\u6bb5\u9762\u5bf9\u7684\u5907\u62e9\u60c5\u666f\u5374\u672a\u5fc5\u6709\u7c7b\u4f3c\u7684\u5bf9\u79f0\u6027[1]\u3002\u4ecd\u7136\u4ee5\u7535\u529b\u884c\u4e1a\u9762\u4e34\u7684\u95ee\u9898\u4e3a\u4f8b\uff0c\u5982\u679c\u9884\u6d4b\u503c\u504f\u9ad8\uff0c\u53ef\u80fd\u5907\u62e9\u65b9\u6848\u4f1a\u589e\u52a0\u4e0d\u5fc5\u8981\u7684\u64cd\u4f5c\u8d44\u6e90\uff0c\u5bfc\u81f4\u65e0\u6548\u7684\u8d44\u6e90\u5206\u914d\u3002\u800c\u5982\u679c\u9884\u6d4b\u503c\u504f\u4f4e\uff0c\u5907\u62e9\u65b9\u6848\u53ef\u80fd\u4f1a\u89e6\u53d1\u7f51\u7edc\u4e4b\u95f4\u7684\u518d\u5e73\u8861\uff0c\u5e26\u6765\u989d\u5916\u7cfb\u7edf\u53ef\u9760\u6027\u95ee\u9898\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n

\u56fe2\uff1a\u7aef\u5230\u7aef\u5b66\u4e60\u65b9\u6cd5\uff08\u8303\u5f0f\u4e8c\uff09<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u7aef\u5230\u7aef\u5b66\u4e60\u65b9\u6cd5\uff08End to End Learning\uff0cE2E\uff09<\/span><\/span><\/strong>\uff1a<\/span><\/span><\/strong>\u4e5f\u79f0\u4e3a\u51b3\u7b56\u9a71\u52a8\u5b66\u4e60\u3002\u8be5\u65b9\u6cd5\u5c06\u4f18\u5316\u7ed3\u6784\u5d4c\u5165\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\uff0c\u901a\u8fc7\u51b3\u7b56\u53cd\u9988\u76f4\u63a5\u8c03\u6574\u9884\u6d4b\u6a21\u578b\uff0c\u5b9e\u73b0\u4ece\u9884\u6d4b\u5230\u51b3\u7b56\u7684\u95ed\u73af\u5b66\u4e60\uff0c\u4ece\u800c\u66f4\u597d\u5730\u5bf9\u9f50\u9884\u6d4b\u4e0e\u51b3\u7b56\u76ee\u6807\uff0c\u6709\u52a9\u4e8e\u51cf\u5c11\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u4e2d\u5b58\u5728\u7684\u504f\u5dee\u95ee\u9898\u3002\u5177\u4f53\u6765\u8bf4\uff0c\u8be5\u65b9\u6cd5\u5c06\u6700\u4f18\u51b3\u7b56\u76ee\u6807<\/span><\/span>\"image.png\"<\/span><\/sub>\u8868\u793a\u4e3a\u9884\u6d4b\u503c\u7684\u51fd\u6570<\/span><\/span>\"image.png\"<\/span><\/span><\/span>\uff0c<\/span><\/span>\u4ece\u800c\u5c06\u51b3\u7b56\u95ee\u9898\u7684\u76ee\u6807\uff08\u516c\u5f0f6-\u516c\u5f0f7\uff09\u8f6c\u5316\u4e3a\u6700\u5c0f\u5316\u4ee5\u4e0b\u635f\u5931\u51fd\u6570\uff08\u4e5f\u79f0\u4e3a\u51b3\u7b56\u95ee\u9898\u7684\u540e\u6094\u51fd\u6570\uff09\uff1a<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n


<\/span><\/span><\/span><\/p>\n

\u4e8e\u662f\u53ef\u4ee5\u8fdb\u4e00\u6b65\u901a\u8fc7\u8ba1\u7b97\uff088\uff09\u7684\u68af\u5ea6\uff0c\u5229\u7528\u68af\u5ea6\u53cd\u5411\u4f20\u64ad\u7b49\u65b9\u6cd5\u5f97\u5230\u6700\u4f18\u51b3\u7b56\u89e3\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n


<\/span><\/span><\/span><\/p>\n

\u4f46\u662f\uff0c\uff089\uff09\u5f0f\u5f80\u5f80\u96be\u4ee5\u663e\u5f0f\u8868\u8fbe\uff0c\u6216\u8005\u5728\u5b9e\u9645\u95ee\u9898\u4e2d\u8ba1\u7b97\u91cf\u5de8\u5927\u3002\u4e00\u822c\u53ef\u4ee5\u901a\u8fc7\u4e09\u7c7bE2E\u65b9\u6cd5\u6c42\u89e3\uff1a<\/span><\/span>\u9690\u51fd\u6570\u65b9\u6cd5<\/span><\/span><\/strong>\uff08Implicit differentiation methods\uff09<\/span><\/span><\/strong>\u3001<\/span><\/span>\u4ee3\u7406\u635f\u5931\u51fd\u6570\u6cd5<\/span><\/span><\/strong>\uff08Surrogate loss methods\uff09<\/span><\/span><\/strong>\u3001<\/span><\/span>\u8fd1\u4f3c\u6c42\u89e3\u6cd5<\/span><\/span><\/strong>\uff08Approximation methods\uff09<\/span><\/span><\/strong>\u3002\u9690\u51fd\u6570\u65b9\u6cd5\u901a\u8fc7\u4f18\u5316\u95ee\u9898\u7684KKT\u6761\u4ef6\u5c06\uff089\uff09\u5f0f\u8868\u793a\u6210\u96c5\u53ef\u6bd4\u77e9\u9635\u7684\u76f8\u5e94\u5f62\u5f0f\u3002\u4ee3\u7406\u635f\u5931\u51fd\u6570\u6cd5\u4ee5SPO+\u635f\u5931\u51fd\u6570\u4e3a\u4ee3\u8868\uff0c\u5c06\u516c\u5f0f\uff089\uff09\u8f6c\u5316\u4e3a\u66f4\u6613\u6c42\u89e3\u7684SPO+\u635f\u5931\u51fd\u6570\u7684\u68af\u5ea6\u5f62\u5f0f[2]\u3002\u8fd1\u4f3c\u6c42\u89e3\u6cd5\u901a\u8fc7\u5bf9\u9884\u6d4b\u503c<\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span><\/span><\/span><\/span>\u53e0\u52a0\u9ad8\u65af\u566a\u58f0<\/span><\/span>\u03f5Z<\/span><\/span><\/span><\/span>\uff0c\u5e76\u8ba1\u7b97\u5176\u671f\u671b\u503c\uff0c\u5373\u51b3\u7b56\u95ee\u9898\u7684\u6270\u52a8\u89e3<\/span><\/span>\"image.png\"<\/span><\/sub>\u3002<\/span><\/span><\/span><\/span><\/p>\n


<\/span><\/p>\n

\u5373\u4fbf\u5982\u6b64\uff0cE2E\u65b9\u6cd5\u4ecd\u7136\u53d7\u5230\u4f18\u5316\u95ee\u9898\uff08\u516c\u5f0f 6-7\uff09\u8fd9\u4e00\u6846\u67b6\u7684\u9650\u5236\uff0c\u5728\u4e00\u4e9b\u95ee\u9898\u4e0a\u7684\u8fed\u4ee3\u8ba1\u7b97\u4efb\u52a1\u4ecd\u7136\u7e41\u91cd\u3002\u56e0\u6b64\u51fa\u73b0\u4e86\u901a\u8fc7\u5728\u7ebf\u6df1\u5ea6\u5f3a\u5316\u5b66\u4e60[3,4]\u3001\u7f51\u683c\u641c\u7d22\u3001\u968f\u673a\u641c\u7d22\u3001\u8d1d\u53f6\u65af\u4f18\u5316[5]\u6765\u51cf\u5c11\u8ba1\u7b97\u91cf\u7684\u65b9\u6cd5\u3002\u53e6\u5916\uff0cE2E\u65b9\u6cd5\u7684<\/span><\/span>\u6cdb\u5316\u80fd\u529b\uff08Transferability\uff09<\/span><\/span><\/strong>\u4e5f\u662f\u4e2a\u95ee\u9898\u3002\u8fd9\u4e3b\u8981\u7531\u4e8eE2E\u65b9\u6cd5\u727a\u7272\u4e86\u9884\u6d4b\u6027\u80fd\uff0c\u6765\u4fdd\u8bc1\u51b3\u7b56\u95ee\u9898\u7684\u4f18\u5316\u6027\u80fd\uff0c\u56e0\u6b64\u5176\u9884\u6d4b\u503c\u53ef\u80fd\u4e0e\u771f\u5b9e\u503c\u504f\u79bb\u8f83\u5927<\/span>\uff08\u56fe2-C\u3001D\uff09<\/span>\uff0c\u4e5f\u5bfc\u81f4\u5176\u6cdb\u5316\u80fd\u529b\u4e0d\u4f73\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n

\u56fe3\uff1a\u76f4\u63a5\u5b66\u4e60\u65b9\u6cd5\uff08\u8303\u5f0f\u4e09\uff09<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u76f4\u63a5\u5b66\u4e60\u65b9\u6cd5\uff08Direct Learning\uff0cDL\uff09<\/span><\/span><\/strong>\u5219\u53e6\u8f9f\u8e4a\u5f84\uff0c\u4e0d\u53bb\u6c42\u89e3\u4f18\u5316\u95ee\u9898\uff08\u516c\u5f0f 6-7\uff09\uff0c\u8f6c\u800c\u5bfb\u627e\u8f93\u5165\u53d8\u91cfx<\/span><\/span>\u4e0e\u51b3\u7b56\u53d8\u91cfz<\/span><\/span>\u7684\u76f4\u63a5\u5173\u7cfb\uff0c\u8be5\u65b9\u6cd5\u4e3b\u8981\u9002\u7528\u4e8e\u90a3\u4e9b\u4f18\u5316\u95ee\u9898\u7ed3\u6784\u590d\u6742\u6216\u9690\u5f0f\u8868\u8fbe\u7684\u573a\u666f\uff0c\u5728\u96be\u4ee5\u660e\u786e\u5b9a\u4e49\u4f18\u5316\u8fc7\u7a0b\u65f6\u8868\u73b0\u51fa\u5353\u8d8a\u7684\u9002\u5e94\u6027\u548c\u7075\u6d3b\u6027\u3002\u4f8b\u5982\uff0c<\/span><\/span>\u5728\u5f3a\u5316\u5b66\u4e60\uff08Reinforcement Learning\uff0cRL\uff09<\/span><\/span><\/strong>\u6846\u67b6\u4e0b\uff0c\u4e3b\u8981\u76ee\u6807\u662f\u6c42\u89e3<\/span><\/span>\u9a6c\u5c14\u53ef\u592b\u51b3\u7b56\u8fc7\u7a0b\uff08Markov Decision Process\uff0cMDP\uff09<\/span><\/span><\/strong>\u7684\u5956\u52b1\u503c\u6700\u5927\u5316\uff0c<\/span><\/span>\"image.png\"<\/span><\/sub>\uff0c<\/span><\/span>\"image.png\"<\/span><\/sub>\uff0c\u03b3<\/span><\/span>\u662f\u6298\u73b0\u7cfb\u6570\uff0c<\/span><\/span><\/span><\/span>r<\/span><\/span>t<\/span><\/sub>(<\/span>x<\/span><\/span>t<\/span><\/sub>, <\/span>z<\/span><\/span>t<\/span><\/sub>) <\/span><\/span><\/span><\/span><\/span><\/span><\/span>\u662f\u5956\u52b1\u51fd\u6570\u3002\u53c8\u5982\u5728<\/span><\/span>\u6a21\u4eff\u5b66\u4e60\uff08Imitation Learning\uff0cIL\uff09<\/span><\/span><\/strong>\u6846\u67b6\u4e0b\uff0c\u4e3b\u8981\u76ee\u6807\u662f\u6c42\u89e3\u7f51\u7edc\u53c2\u6570<\/span><\/span>\u03a6\uff0c\u4f7f\u5f97\u6a21\u4eff\u7f51\u7edc\u7684\u884c\u4e3a\u4e0e\u4e13\u5bb6\u793a\u8303\u6570\u636e\u5c3d\u53ef\u80fd\u4e00\u81f4\uff0c<\/span><\/span>\"image.png\"<\/span><\/sub><\/span><\/span>\uff0c\u8fd9\u91cc\u7684<\/span><\/span>y<\/span><\/span><\/span><\/span>\u8868\u793a\u4e13\u5bb6\u793a\u8303\u6570\u636e\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n

\u56fe4\uff1a\u4e09\u79cd\u6570\u636e\u9a71\u52a8\u4f18\u5316\u8303\u5f0f\u4e4b\u95f4\u7684\u6bd4\u8f83<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u4e09\u7c7b\u65b9\u6cd5\u5747\u6709\u4e0d\u540c\u7684\u7279\u70b9\u548c\u9002\u7528\u8303\u56f4\uff0c\u7b80\u5355<\/span><\/span>\u5bf9\u6bd4\u603b\u7ed3<\/span><\/span><\/strong>\u4e3a\uff1a<\/span><\/span><\/p>\n


<\/span><\/p>\n

1.\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u7684\u53ef\u89e3\u91ca\u6027\u5f3a\u3001\u8ba1\u7b97\u6709\u6548\u6027\u8f83\u597d\uff0c\u9002\u7528\u4e8e\u5bf9\u9884\u6d4b\u8981\u6c42\u8f83\u9ad8\uff0c\u9884\u6d4b\u6a21\u578b\u91cd\u8bad\u7ec3\u96be\u5ea6\u8f83\u5927\u7684\u5927\u578b\u95ee\u9898<\/span>\uff08\u5982NLP\u3001\u6c14\u8c61\u95ee\u9898\uff09<\/span>\uff0c\u540c\u65f6\u53ef\u4ee5\u63a5\u53d7\u4e00\u5b9a\u7a0b\u5ea6\u7684\u6b21\u4f18\u51b3\u7b56\uff1b<\/span><\/span><\/p>\n


<\/span><\/p>\n

2.\u7aef\u5230\u7aef\u5b66\u4e60\u65b9\u6cd5\u7684\u4e00\u81f4\u6027\u5f3a\u3001\u53ef\u89e3\u91ca\u6027\u597d\uff0c\u9002\u7528\u4e8e\u51b3\u7b56\u7cbe\u5ea6\u8981\u6c42\u9ad8\u7684\u60c5\u666f\uff1b<\/span><\/span><\/p>\n


<\/span><\/p>\n

3.\u76f4\u63a5\u5b66\u4e60\u65b9\u6cd5\u7684\u4e00\u81f4\u6027\u5f3a\u3001\u8ba1\u7b97\u6709\u6548\u6027\u76f8\u5bf9\u8f83\u597d\uff0c\u9002\u7528\u4e8e\u590d\u6742\u3001\u52a8\u6001\u7684\u51b3\u7b56\u73af\u5883\uff0c\u6bd4\u5982\u81ea\u52a8\u9a7e\u9a76\u9886\u57df\u7b49\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u867d\u7136\u4e09\u7c7b\u65b9\u6cd5\u5747\u4e3a\u6570\u636e\u9a71\u52a8\u7684\u4f18\u5316\u65b9\u6cd5\uff0c\u4f46\u662f\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u4e00\u822c\u4f5c\u4e3a\u57fa\u7ebf\u65b9\u6cd5\u7684\u89d2\u8272\u51fa\u73b0\u3002\u5728\u76ee\u524d\u5927\u90e8\u5206\u7684\u6570\u636e\u9a71\u52a8\u5e94\u7528\u573a\u666f\u91cc\uff0c\u6bd4\u5982\u7535\u529b\u7f51\u7edc\u89c4\u5212\u3001\u590d\u6742\u7cfb\u7edf\u4f18\u5316\u3001\u80fd\u6e90\u7cfb\u7edf\u8282\u80fd\u3001\u8d27\u7269\u5feb\u9012\u3001\u6e2f\u53e3\u7ba1\u7406\u3001\u81ea\u52a8\u9a7e\u9a76\u7b49\uff0c\u7aef\u5230\u7aef\u5b66\u4e60\u65b9\u6cd5\u548c\u76f4\u63a5\u5b66\u4e60\u65b9\u6cd5\u5728\u6548\u679c\u4e0a\u90fd\u8981\u80dc\u8fc7\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u3002<\/span><\/span><\/p>\n


<\/span><\/h2>\n


<\/span><\/h2>\n

\n
\n
\n
\n
\n

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

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

\u4ece\u7406\u8bba\u5230\u5e94\u7528\uff1a<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n

\u6570\u636e\u9a71\u52a8\u4f18\u5316\u65b9\u6cd5\u9762\u4e34\u7684\u6311\u6218<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

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

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


<\/span><\/p>\n

\u76f8\u8f83\u4e8e\u57fa\u4e8e\u5148\u9a8c\u77e5\u8bc6\u548c\u6a21\u578b\u7684\u4f20\u7edf\u51b3\u7b56\u65b9\u6cd5\uff0c\u6570\u636e\u4e0e\u9884\u6d4b\u6280\u672f\u7684\u878d\u5408\u663e\u8457\u63d0\u5347\u4e86\u5b9e\u9645\u5e94\u7528\u6548\u80fd\u3002\u4f8b\u5982\uff0c\u901a\u8fc7\u6574\u5408\u5927\u89c4\u6a21\u6c14\u8c61\u6570\u636e\u4e0e\u5730\u7406\u4fe1\u606f\uff0c\u53ef\u5b9e\u73b0\u66f4\u7cbe\u51c6\u9ad8\u6548\u7684\u53ef\u518d\u751f\u80fd\u6e90\u9884\u6d4b\uff0c\u8fdb\u800c\u501f\u52a9\u591a\u5c3a\u5ea6\u4eba\u5de5\u667a\u80fd\u6a21\u578b\u4f7f\u5fae\u7535\u7f51\u4e2d\u7684\u5206\u5e03\u5f0f\u53d1\u7535\u66f4\u667a\u80fd\u9ad8\u6548\u3002\u6b64\u5916\uff0c\u9488\u5bf9\u53d1\u7535\u4f01\u4e1a\u4e0e\u7535\u7f51\u8fd0\u8425\u5546\u666e\u904d\u9762\u4e34\u7684\u7ba1\u7406\u96be\u9898\uff0cIBM\u3001Oracle\u7b49\u884c\u4e1a\u5de8\u5934\u5df2\u5f00\u53d1\u51fa\u5927\u6570\u636e\u9a71\u52a8\u7684\u667a\u6167\u80fd\u6e90\u7ba1\u7406\u4ea7\u54c1\u4e0e\u670d\u52a1\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u8d44\u4ea7\u5bc6\u96c6\u578b\u4ea7\u4e1a\u3002\u603b\u4f53\u800c\u8a00\uff0c\u6570\u636e\u8f93\u5165\u4e0e\u9884\u6d4b\u5206\u6790\u7b49\u6280\u672f\u7684\u51fa\u73b0\u4e3a\u51b3\u7b56\u5e94\u7528\u6ce8\u5165\u4e86\u65b0\u6d3b\u529b\uff0c\u63a8\u52a8\u884c\u4e1a\u5b9e\u8df5\u4e0e\u6807\u51c6\u5b9e\u73b0\u8d28\u7684\u98de\u8dc3\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u6587\u7ae0\u901a\u8fc7\u4e09\u4e2a\u5fae\u578b\u6848\u4f8b\u7814\u7a76\uff0c\u5206\u522b\u5c55\u793a\u6570\u636e\u9a71\u52a8\u4f18\u5316\u65b9\u6cd5\u5728\u7535\u7f51\u8c03\u5ea6\u3001\u5e02\u573a\u8fd0\u8425\u53ca\u65e0\u4eba\u81ea\u52a8\u63a7\u5236\u7b49\u4e0d\u540c\u9886\u57df\u7684\u591a\u5143\u5316\u5e94\u7528\u3002\u8fd9\u4e9b\u6848\u4f8b\u51f8\u663e\u4e86\u5404\u7c7b\u6280\u672f\u5728\u5e94\u5bf9\u590d\u6742\u51b3\u7b56\u6311\u6218\u65f6\u5177\u6709\u7684\u8de8\u9886\u57df\u9002\u7528\u6027\u4e0e\u5b9e\u6548\u6027\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u7535\u529b<\/span><\/span><\/strong>\u8c03\u5ea6\uff1a<\/span><\/span><\/strong>Wahdany\u7b49\u4eba[6]\u5229\u7528KKT\u6761\u4ef6<\/span> (Karush-Kuhn-Tucker conditions) <\/span>\u6765\u89e3\u51b3\u9ad8\u98ce\u7535\u7535\u7f51\u7684\u6f6e\u6d41\u63a7\u5236\u95ee\u9898\uff0c\u4ed6\u4eec\u4f7f\u7528\u7684\u7aef\u5230\u7aef\u5b66\u4e60\u65b9\u6cd5\u8981\u6bd4\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u51cf\u5c11\u4e868.5%\u7684\u7f51\u7edc\u62e5\u5835\u548c\u8fc7\u8f7d\u95ee\u9898\u3002Zhou\u7b49\u4eba[7]\u4f7f\u7528LSTM\u9884\u6d4b\u65b9\u6cd5\u6c42\u89e3\u7535\u7f51\u4e2d\u7684\u6df7\u5408\u7ebf\u6027\u89c4\u5212<\/span>\uff08Mixed-Integer Linear Programming\uff0c MILP\uff09<\/span>\u95ee\u9898\uff0c\u964d\u4f4e\u4e860.4%\u7684\u8fd0\u8425\u6210\u672c\u3002Sang\u7b49\u4eba[8]\u4f7f\u7528SPO+\u635f\u5931\u51fd\u6570\uff0c\u6839\u636e\u7535\u4ef7\u6765\u4f18\u5316\u50a8\u80fd\u7cfb\u7edf\u884c\u4e3a\uff0c\u6bd4\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u63d0\u9ad8\u4e866.11%\u7684\u7ecf\u6d4e\u6548\u76ca\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u5e02\u573a\u8fd0\u8425\uff1a<\/span><\/span><\/strong>Chu\u7b49\u4eba[9]\u5229\u7528SPO+\u65b9\u6cd5\u89e3\u51b3\u672b\u7aef\u9001\u8d27\u95ee\u9898\uff0c\u4e0e\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u76f8\u6bd4\u51cf\u5c11\u4e865%\u7684\u8fd0\u8f93\u6210\u672c\u3002Tian\u7b49\u4eba[10]\u5728\u7aef\u5230\u7aef\u5b66\u4e60\u6846\u67b6\u4e0b\u6c42\u89e3\u6e2f\u53e3\u5de1\u68c0\u95ee\u9898\uff0c\u6bd4\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u51cf\u5c11\u6bcf\u8258\u8239\u53ea1%\u7684\u8fd0\u8425\u6210\u672c\u3002Qi\u7b49\u4eba[11]\u63d0\u51fa\u4e86\u4e0d\u786e\u5b9a\u9700\u6c42\u4e0b\u7684\u5e93\u5b58\u7ba1\u7406\u7684\u76f4\u63a5\u5b66\u4e60\u6846\u67b6\uff0c\u6bd4\u5e8f\u8d2f\u4f18\u5316\u65b9\u6cd5\u5e73\u5747\u964d\u4f4e26.1%\u7684\u6301\u6709\u6210\u672c<\/span>[\u6ce8\uff1a\u5305\u62ec\u8d44\u91d1\u6210\u672c\u3001\u4ed3\u50a8\u6210\u672c\u3001\u4fdd\u9669\u6210\u672c\u548c\u5e93\u5b58\u8d2c\u503c\u6210\u672c\uff0c\u53cd\u6620\u4e86\u6301\u6709\u5e93\u5b58\u6240\u5e26\u6765\u7684\u4e00\u7cfb\u5217\u8d39\u7528]<\/span>\u548c51.7%\u7684\u7f3a\u8d27\u6210\u672c<\/span>[\u6ce8\uff1a\u5305\u62ec\u9500\u552e\u635f\u5931\u3001\u5ba2\u6237\u6d41\u5931\u6210\u672c\u3001\u7d27\u6025\u91c7\u8d2d\u6210\u672c\u548c\u751f\u4ea7\u4e2d\u65ad\u6210\u672c\uff0c\u53cd\u6620\u4e86\u56e0\u5e93\u5b58\u4e0d\u8db3\u800c\u5e26\u6765\u7684\u635f\u5931]<\/span>\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u65e0\u4eba\u81ea\u52a8\u63a7\u5236<\/span><\/span><\/strong>\uff1a<\/span><\/span><\/strong>\u5728\u81ea\u52a8\u9a7e\u9a76\u3001\u65e0\u4eba\u673a\u63a7\u5236\u3001\u673a\u5668\u4eba\u9886\u57df\uff0c\u7531\u4e8e\u73af\u5883\u548c\u63a7\u5236\u4efb\u52a1\u7684\u590d\u6742\uff0c\u51b3\u7b56\u76ee\u6807\u7684\u4f18\u5316\u95ee\u9898\u7684\u663e\u5f0f\u8868\u8fbe\u51e0\u65e0\u53ef\u80fd\u3002\u8fd9\u7c7b\u95ee\u9898\u5e7f\u6cdb\u4f7f\u7528\u76f4\u63a5\u5b66\u4e60\u65b9\u6cd5\u6c42\u89e3\u3002\u6bd4\u5982\uff0cCao\u7b49\u4eba[12]\u5f15\u5165\u52a8\u6001\u7f6e\u4fe1\u611f\u77e5\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff0c\u6765\u5904\u7406\u81ea\u52a8\u9a7e\u9a76\u8f66\u8f86\u5728\u6781\u7aef\u6761\u4ef6\u4e0b\u7684\u4e0d\u786e\u5b9a\u6027\u95ee\u9898\u3002Kim\u7b49\u4eba[13]\u4f7f\u7528\u5f3a\u5316\u5b66\u4e60\u65b9\u6cd5\u8bad\u7ec3\u65e0\u4eba\u673a\uff0c\u6574\u5408\u65e0\u4eba\u673a\u7684\u5f62\u53d8\u4f20\u611f\u6570\u636e\u3001\u59ff\u6001\u6570\u636e\u3001\u98ce\u5411\u98ce\u901f\u6570\u636e\uff0c\u6765\u5b9e\u73b0\u6700\u4f18\u98de\u884c\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u9488\u5bf9\u6570\u636e\u9a71\u52a8\u4f18\u5316\u7406\u8bba\u5728\u884c\u4e1a\u5e94\u7528\u60c5\u51b5\uff0c\u6587\u7ae0\u4ece\u201c\u6570\u636e\u8d28\u91cf\u2013\u5efa\u6a21\u4f18\u5316\u2013\u51b3\u7b56\u5e94\u7528\u201d\u7684\u5168\u6d41\u7a0b\u5206\u6790\u6311\u6218\u5e76\u63d0\u51fa\u76f8\u5e94\u6280\u672f\u6027\u89e3\u51b3\u65b9\u6848\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n

\u56fe5\uff1a\u6570\u636e\u9a71\u52a8\u4f18\u5316\u65b9\u6cd5\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\u7684\u6311\u6218<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u4e0a\u56fe\u663e\u793a\u4e86\u6570\u636e\u9a71\u52a8\u4f18\u5316\u5728\u5b9e\u9645\u5e94\u7528\u573a\u666f\u4e2d\u9762\u4e34\u7684\u4e09\u5927\u6311\u6218\uff0c\u5177\u4f53\u6765\u8bf4\u5305\u62ec\u4ee5\u4e0b\u4e09\u4e2a\u65b9\u9762\uff1a<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u6570\u636e\u7684\u8d28\u91cf\u548c\u6570\u91cf\u51b3\u5b9a\u4e86\u51b3\u7b56\u7684\u6548\u679c\u548c\u6548\u7387\u3002<\/span><\/span><\/strong>\u4ece\u6570\u636e\u8d28\u91cf\u6765\u770b\uff0c\u6570\u636e\u9a71\u52a8\u65b9\u6cd5\u8981\u6c42\u5bf9\u6570\u636e\u8fdb\u884c\u826f\u597d\u7684\u9884\u5904\u7406\u3002\u4f20\u7edf\u7684\u65e0\u76d1\u7763\u7684\u6570\u636e\u9884\u5904\u7406\u65b9\u6cd5\u5305\u62ec\u4e00\u81f4\u6027\u5206\u6790[14]\u3001\u9999\u519c\u71b5[15]\u3001\u4f8b\u5916\u70b9\u76d1\u6d4b[16]\uff0c\u4f46\u662f\u7531\u4e8e\u51b3\u7b56\u95ee\u9898\u7684\u590d\u6742\u6027\uff0c\u6570\u636e\u7684\u91cd\u8981\u6027\u5f88\u53ef\u80fd\u662f\u4e0e\u51b3\u7b56\u6761\u4ef6\u76f8\u5173\u7684\u3002\u4e3a\u6b64\uff0c<\/span><\/span>\u9700\u8981\u5f15\u5165\u5982\u52a8\u6001\u5206\u5e03\u65b9\u6cd5[17]\u4e4b\u7c7b\u7684\u6709\u76d1\u7763\u7684\u6570\u636e\u9884\u5904\u7406\u65b9\u6cd5\uff0c\u6765\u589e\u52a0\u51b3\u7b56\u9636\u6bb5\u7684\u6570\u636e\u8d28\u91cf<\/span><\/span><\/strong>\u3002\u8fd9\u65b9\u9762\u7684\u7814\u7a76\u4ecd\u7136\u5904\u4e8e\u76f8\u5bf9\u65e9\u671f\u9636\u6bb5\u3002\u4ece\u6570\u636e\u6570\u91cf\u6765\u770b\uff0c\u6570\u636e\u91cf\u592a\u5c11\u53ef\u80fd\u9020\u6210\u8fc7\u62df\u5408\u95ee\u9898\uff0c\u592a\u591a\u5219\u53ef\u80fd\u589e\u52a0\u8bad\u7ec3\u6210\u672c\u548c\u8ba1\u7b97\u65f6\u95f4\u3002\u5728\u6570\u636e\u6570\u91cf\u548c\u51b3\u7b56\u6548\u679c\u4e4b\u95f4\u53d6\u5f97\u5e73\u8861\uff0c\u4ecd\u7136\u662f\u4e00\u4e2a\u5173\u952e\u7684\u7814\u7a76\u8bfe\u9898\u3002\u5bf9\u5927\u6837\u672c\u800c\u8a00\uff0c\u9700\u8981\u786e\u5b9a\u6700\u4f18\u91c7\u6837\u5ea6[18]\uff0c\u5bf9\u5c0f\u6837\u672c\u800c\u8a00\uff0c\u5219\u9700\u8981\u5f15\u5165\u6570\u636e\u589e\u5e7f\u7b49\u6280\u672f[19,20]\uff0c\u4ee5\u53ca\u5c0f\u6837\u672c\u5b66\u4e60\u65b9\u6cd5[21]\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u4e0d\u786e\u5b9a\u6027\u5efa\u6a21\u548c\u89e3\u7684\u53ef\u884c\u6027\u51b3\u5b9a\u4e86\u4f18\u5316\u65b9\u6cd5\u7684\u6548\u80fd\u3002<\/span><\/span><\/strong>\u5728\u4e0d\u786e\u5b9a\u6027\u5efa\u6a21\u65b9\u9762\uff0c\u65b9\u6cd5\u7684\u5e94\u7528\u5148\u4e8e\u7406\u8bba\u6307\u5bfc\u3002\u8fd1\u5341\u5e74\u6765\uff0c\u6d8c\u73b0\u4e86\u57fa\u4e8e\u671f\u671b\u7684\u9690\u51fd\u6570\u6cd5[22]\u3001\u968f\u673a\u7ea6\u675f[23]\u3001\u7a33\u5065\u4f18\u5316[24,25]\u7b49\u65b9\u6cd5\uff0c\u4f46\u662f\u4e0d\u786e\u5b9a\u6027\u5efa\u6a21\u7684\u7406\u8bba\u5206\u6790\u4ecd\u7136\u6ede\u540e\u3002\u5728\u89e3\u7684\u53ef\u884c\u6027\u65b9\u9762\uff0c\u6700\u5927\u7684\u6311\u6218\u6765\u81ea\u4e8e\u5982\u4f55\u5728E2E\u7b49\u65b9\u6cd5\u5e94\u7528\u8fc7\u7a0b\u4e2d\uff0c\u4ecd\u7136\u4fdd\u6301\u89e3\u7684\u53ef\u884c\u6027\u3002\u867d\u7136\u5982\u589e\u52a0\u60e9\u7f5a\u9879[26]\u3001\u5f15\u5165\u98ce\u9669\u9879[27]\u3001\u52a8\u6001\u8c03\u6574\u7ea6\u675f\u9879[22]\u7b49\u65b9\u6cd5\u80fd\u5728\u4e00\u5b9a\u7a0b\u5ea6\u4e0a\u4fdd\u8bc1\u53ef\u884c\u89e3\uff0c\u4f46\u662f\u5b8c\u6574\u7684\u7406\u8bba\u5206\u6790\u6846\u67b6\u4ecd\u7136\u6709\u5f85\u5efa\u7acb\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n

\u53ef\u89e3\u91ca\u6027\u548c\u53ef\u62d3\u5c55\u6027\u51b3\u5b9a\u4e86\u51b3\u7b56\u65b9\u6cd5\u7684\u5e94\u7528\u8303\u56f4\u3002<\/span><\/span><\/strong>\u53ef\u89e3\u91ca\u6027\u4e00\u65b9\u9762\u80fd\u591f\u589e\u5f3a\u51b3\u7b56\u8005\u5bf9\u51b3\u7b56\u65b9\u6cd5\u7684\u4fe1\u5fc3\uff0c\u53e6\u4e00\u65b9\u9762\u8fd8\u6709\u52a9\u4e8e\u589e\u52a0\u51b3\u7b56\u8005\u5bf9\u51b3\u7b56\u95ee\u9898\u7684\u6d1e\u89c1\u3002\u673a\u5668\u5b66\u4e60\u9886\u57df\u51fa\u73b0\u4e86\u4e00\u4e9b\u6709\u6548\u7684\u89e3\u91ca\u6027\u65b9\u6cd5\uff0c\u5305\u62ecLIME<\/span>\uff08Local Interpretable Model-agnostic Explanations\uff09<\/span>\uff0c\u901a\u8fc7\u5c40\u90e8\u8fd1\u4f3c\u7684\u65b9\u5f0f\uff0c\u4e3a\u590d\u6742\u7684\u201c\u9ed1\u7bb1\u201d\u6a21\u578b\u63d0\u4f9b\u53ef\u89e3\u91ca\u6027\uff0c\u5e2e\u52a9\u7528\u6237\u7406\u89e3\u6a21\u578b\u5728\u7279\u5b9a\u9884\u6d4b\u4e0a\u7684\u51b3\u7b56\u4f9d\u636e\u3002\u4ee5\u53ca SHAP<\/span>\uff08SHapley Additive exPlanations\uff09<\/span>\uff0c\u5229\u7528\u516c\u5e73\u5206\u914d\u5408\u4f5c\u535a\u5f08\u6536\u76ca\u7684 Shapley Value\uff0c\u8ba1\u7b97\u6bcf\u4e2a\u7279\u5f81\u5bf9\u6a21\u578b\u9884\u6d4b\u7684\u8d21\u732e\uff0c\u5e2e\u52a9\u7528\u6237\u7406\u89e3\u6a21\u578b\u7684\u51b3\u7b56\u8fc7\u7a0b\u3002\u4f46\u662f\u5c06\u89e3\u91ca\u6027\u65b9\u6cd5\u4ece\u673a\u5668\u5b66\u4e60\u9886\u57df\u62d3\u5c55\u5230\u51b3\u7b56\u7406\u8bba\uff0c\u4ecd\u5b58\u5728\u8f83\u5927\u6311\u6218\u3002\u53ef\u62d3\u5c55\u6027\u6307\u6a21\u578b\u5728\u52a8\u6001\u73af\u5883\u3001\u4e0d\u7a33\u5b9a\u5206\u5e03\u6761\u4ef6\u4e0b\u4ecd\u7136\u80fd\u53d6\u5f97\u826f\u597d\u6548\u679c\u7684\u80fd\u529b\uff0c\u662f\u6a21\u578b\u80fd\u591f\u8de8\u9886\u57df\u8de8\u884c\u4e1a\u5e94\u7528\u7684\u91cd\u8981\u8861\u91cf\u3002\u6a21\u578b\u4e0d\u53ef\u77e5\u7684\u5143\u5b66\u4e60\u65b9\u6cd5[28]\u3001\u5728\u7ebf\u5b66\u4e60\u65b9\u6cd5[29]\u3001\u5c0f\u6837\u672c\u5b66\u4e60\u65b9\u6cd5[30]\u90fd\u662f\u4fdd\u8bc1\u53ef\u62d3\u5c55\u6027\u7684\u91cd\u8981\u65b9\u6cd5\u7c7b\u578b\u3002<\/span><\/span><\/p>\n


<\/span><\/h2>\n


<\/span><\/h2>\n

\n
\n
\n
\n
\n

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

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

\u7ed3\u8bba<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

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

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


<\/span><\/p>\n

\u5927\u6570\u636e\u6df1\u523b\u7684\u6539\u53d8\u7740\u4eba\u4eec\u7684\u751f\u6d3b\uff0c\u6570\u636e\u79d1\u5b66\u540c\u6837\u6df1\u523b\u7684\u6539\u53d8\u7740\u4eba\u4eec\u7684\u51b3\u7b56\u884c\u4e3a\u3002\u5728\u6570\u636e\u9a71\u52a8\u7684\u4f18\u5316\u65b9\u6cd5\u63a8\u52a8\u4e0b\uff0c\u51b3\u7b56\u95ee\u9898\u7684\u6c42\u89e3\u5c06\u53d8\u5f97\u8d8a\u6765\u8d8a\u4fbf\u6377\u3001\u5feb\u901f\u3002\u4eceNexus\u7684\u8fd9\u7bc7\u8bba\u6587\u53ef\u4ee5\u770b\u5230\uff0c<\/span><\/span>\u6570\u636e\u4f18\u5316\u7684\u9a71\u52a8\u65b9\u6cd5\u7ecf\u5386\u4e86\u4e09\u6b21\u66f4\u8fed\uff0c\u4ece<\/span><\/span><\/strong>\u5e8f\u8d2f\u4f18\u5316<\/span><\/span><\/strong>\u5230<\/span><\/span><\/strong>\u7aef\u5230\u7aef\u5b66\u4e60\u4f18\u5316<\/span><\/span><\/strong>\u518d\u5230<\/span><\/span><\/strong>\u76f4\u63a5\u5b66\u4e60\u65b9\u6cd5<\/span><\/span><\/strong>\uff0c\u6a21\u578b\u7684\u5185\u5728\u4e00\u81f4\u6027\u66f4\u5f3a\uff0c\u8ba1\u7b97\u6548\u7387\u66f4\u9ad8<\/span><\/span><\/strong>\u3002\u867d\u7136\uff0c\u8fd9\u4e9b\u65b9\u6cd5\u5728\u6570\u636e\u65b9\u6cd5\u8bba\u3001\u4e0d\u786e\u5b9a\u6027\u3001\u89e3\u7684\u53ef\u884c\u6027\u3001\u53ef\u89e3\u91ca\u6027\u548c\u53ef\u62d3\u5c55\u6027\u65b9\u9762\u90fd\u5b58\u5728\u6311\u6218\uff0c\u4f46\u968f\u7740\u6570\u636e\u79d1\u5b66\u3001\u4f18\u5316\u7406\u8bba\u7684\u53d1\u5c55\uff0c\u65b0\u7684\u51b3\u7b56\u65b9\u6cd5\u5c06\u5728\u66f4\u5e7f\u9614\u7684\u9886\u57df\u5f97\u5230\u5e94\u7528\u3002\u6570\u636e\u9a71\u52a8\u7684\u51b3\u7b56\u79d1\u5b66\u5c06\u662f\u4e00\u4e2a\u6709\u751f\u547d\u529b\u7684\u3001\u65b9\u5174\u672a\u827e\u7684\u7814\u7a76\u9886\u57df\u3002<\/span><\/span><\/p>\n


<\/span><\/p>\n


<\/span><\/p>\n

\n
\n
\n

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

\n
\n
\n
\n


<\/span><\/p>\n

    \n
  1. \n

    Zareipour, H., Canizares, C.A., and Bhattacharya, K. (2010). Economic impact of electricity market price forecasting errors: A demand-side analysis. IEEE Trans. Power Syst. 25, 254-262.<\/span><\/span><\/p>\n<\/li>\n

  2. \n

    Elmachtoub, A. N., & Grigas, P. (2017). Smart \u201cPredict, then Optimize\u201d. arXiv preprint arXiv:1710.08005.<\/span><\/span><\/p>\n<\/li>\n

  3. \n

    Bertsimas, D., and Stellato, B. (2022). Online mixed-integer optimization in milliseconds. Inf. J. Comput. 34, 2229-2248.<\/span><\/span><\/p>\n<\/li>\n

  4. \n

    Tang, Y., Agrawal, S., and Faenza, Y. (2020). Reinforcement learning for integer programming: Learning to cut. In International Conference on Machine Learning (PMLR), pp. 9367-9376.<\/span><\/span><\/p>\n<\/li>\n

  5. \n

    Feurer, M., and Hutter, F. (2019). Hyperparameter optimization. In Automated Machine Learning: Methods, Systems, Challenges (Springer International Publishing), pp. 3-33.<\/span><\/span><\/p>\n<\/li>\n

  6. \n

    Wahdany, D., Schmitt, C., and Cremer, J.L. (2023). More than accuracy: end-to-end wind power forecasting that optimises the energy system. Elec. Power Syst. Res. 221, 109384.<\/span><\/span><\/p>\n<\/li>\n

  7. \n

    Zhou, Y., Wen, Q., Song, J., et al. (2024). Load data valuation in multienergy systems: An end-to-end approach. IEEE Trans. Smart Grid.<\/span><\/span><\/p>\n<\/li>\n

  8. \n

    Sang, L., Xu, Y., Long, H., et al. (2022). Electricity price prediction for energy storage system arbitrage: A decision-focused approach. IEEE Trans. Smart Grid 13, 2822-2832.<\/span><\/span><\/p>\n<\/li>\n

  9. \n

    Chu, H., Zhang, W., Bai, P., et al. (2023). Data-driven optimization for last-mile delivery. Complex Intell. Systems 9, 2271-2284.<\/span><\/span><\/p>\n<\/li>\n

  10. \n

    Tian, X., Yan, R., Liu, Y., et al. (2023). A smart predict-then-optimize method for targeted and cost-effective maritime transportation. Transp. Res. Part B Methodol. 172, 32-52.<\/span><\/span><\/p>\n<\/li>\n

  11. \n

    Qi, M., Shi, Y., Qi, Y., et al. (2023). A practical end-to-end inventory management model with deep learning. Manag. Sci. 69, 759-773.<\/span><\/span><\/p>\n<\/li>\n

  12. \n

    Cao, Z., Jiang, K., Zhou, W., et al. (2023). Continuous improvement of self driving cars using dynamic confidence-aware reinforcement learning. Nat. Mach. Intell. 5, 145-158.<\/span><\/span><\/p>\n<\/li>\n

  13. \n

    Kim, T., Hong, I., Im, S., et al. (2024). Wing-strain-based flight control of flapping-wing drones through reinforce-ment learning. Nat. Mach. Intell. 6, 992-1005.<\/span><\/span><\/p>\n<\/li>\n

  14. \n

    Cichy, C., and Rass, S. (2019). An overview of data quality frameworks. IEEE Access 7, 24634-24648.<\/span><\/span><\/p>\n<\/li>\n

  15. \n

    Lin, J. (1991). Divergence measures based on the shannon entropy. IEEE Trans. Inf. Theor. 37, 145-151.<\/span><\/span><\/p>\n<\/li>\n

  16. \n

    Bl\u00e1zquez-Garc\u00eda, A., Conde, A., Mori, U., et al. (2021). A review on out-lier\/anomaly detection in time series data. ACM Comput. Surv. 54, 1-33.<\/span><\/span><\/p>\n<\/li>\n

  17. \n

    Ghorbani, A., Kim, M., and Zou, J. (2020). A distributional framework for data valuation. In International Conference on Machine Learning (PMLR), pp. 3535-3544.<\/span><\/span><\/p>\n<\/li>\n

  18. \n

    Besbes, O., and Mouchtaki, O. (2023). How big should your data really be? data driven newsvendor: Learning one sample at a time. Manag. Sci. 69, 5848-5865.<\/span><\/span><\/p>\n<\/li>\n

  19. \n

    Lata, K., Dave, M., and KN, N. (2019). Data augmentation using generative adversarial network. In Proceedings of 2nd International Conference on Advanced Computing and Software Engineering (ICACSE).<\/span><\/span><\/p>\n<\/li>\n

  20. \n

    Mumuni, A., and Mumuni, F. (2022). Data augmentation: A comprehensive survey of modern approaches. Array 16, 100258.<\/span><\/span><\/p>\n<\/li>\n

  21. \n

    Wang, Y., Yao, Q., Kwok, J.T., et al. (2020). Generalizing from a few examples: A survey on few-shot learning. ACM Comput. Surv. 53, 1-34.<\/span><\/span><\/p>\n<\/li>\n

  22. \n

    Donti, P., Amos, B., and Kolter, J.Z. (2017). Task-based end-to-end model learning in stochastic optimization. Adv. Neural Inf. Process. Syst. 30.<\/span><\/span><\/p>\n<\/li>\n

  23. \n

    Pr’ekopa, A. (2013). Stochastic Programming, 324 (Springer Science & Business Media).<\/span><\/span><\/p>\n<\/li>\n

  24. \n

    Rahimian, H., and Mehrotra, S. (2019). Distributionally robust optimization: A review. Preprint at arXiv. https:\/\/doi.org\/10.48550\/arXiv.1908.05659.<\/span><\/span><\/p>\n<\/li>\n

  25. \n

    Ben-Tal, A., El Ghaoui, L., and Nemirovski, A. (2009). Robust Optimization, 28 (Princeton university press).<\/span><\/span><\/p>\n<\/li>\n

  26. \n

    Garcia, J.D., Street, A., Homem-de Mello, T., et al. (2021). Application-driven learning: A closed-loop prediction and optimization approach applied to dynamic reserves and demand forecasting. Preprint at arXiv. https:\/\/doi.org\/10.48550\/arXiv.2102.13273.<\/span><\/span><\/p>\n<\/li>\n

  27. \n

    Zhang, H., Li, R., Chen, Y., et al. (2023). Risk-aware objective-based forecasting in inertia management. IEEE Trans. Power Syst.<\/span><\/span><\/p>\n<\/li>\n

  28. \n

    Finn, C., Abbeel, P., and Levine, S. (2017). Model-agnostic meta-learning for fast adaptation of deep networks. In International Conference on Machine Learning (PMLR), pp. 1126-1135.<\/span><\/span><\/p>\n<\/li>\n

  29. \n

    Hoi, S.C., Sahoo, D., Lu, J., et al. (2021). Online learning: A comprehensive survey. Neurocomputing 459, 249-289.<\/span><\/span><\/p>\n<\/li>\n

  30. \n

    Parnami, A., and Lee, M. (2022). Learning from few examples: A summary of approaches to few-shot learning. Preprint at arXiv. https:\/\/doi.org\/10.48550\/arXiv.2203.04291.<\/span><\/span><\/p>\n<\/li>\n<\/ol>\n


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

    \n
    \n

    \u53c2\u8003\u6587\u732e\u53ef<\/span>\u4e0a\u4e0b\u6ed1\u52a8<\/span><\/strong>\u67e5\u770b<\/span><\/p>\n<\/section>\n<\/section>\n


    <\/span><\/p>\n


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

    \u590d\u6742\u7cfb\u7edf\u81ea\u52a8\u5efa\u6a21\u8bfb\u4e66\u4f1a\u7b2c\u4e8c\u5b63<\/span><\/strong>
    <\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n


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

    \u201c\u590d\u6742\u4e16\u754c\uff0c\u7b80\u5355\u89c4\u5219\u201d\u3002<\/span><\/span><\/p>\n


    <\/span><\/span><\/p>\n

    \u96c6\u667a\u4ff1\u4e50\u90e8\u8054\u5408\u590d\u65e6\u5927\u5b66\u667a\u80fd\u590d\u6742\u4f53\u7cfb\u5b9e\u9a8c\u5ba4\u9752\u5e74\u7814\u7a76\u5458\u6731\u7fa4\u559c\u3001\u6d59\u6c5f\u5927\u5b66\u767e\u4eba\u8ba1\u5212\u7814\u7a76\u5458\u674e\u6a35\u98ce\u3001\u6e05\u534e\u5927\u5b66\u7535\u5b50\u5de5\u7a0b\u7cfb\u6570\u636e\u79d1\u5b66\u4e0e\u667a\u80fd\u5b9e\u9a8c\u5ba4\u535a\u58eb\u540e\u7814\u7a76\u5458\u4e01\u749f\u97ec\u3001\u7f8e\u56fd\u4e1c\u5317\u5927\u5b66\u7269\u7406\u7cfbAlbert-L\u00e1szl\u00f3 Barab\u00e1si\u6307\u5bfc\u7684\u535a\u58eb\u540e\u9ad8\u5a77\u5a77\u3001\u5317\u4eac\u5927\u5b66\u535a\u96c5\u535a\u58eb\u540e\u66f9\u6587\u797a\u3001\u590d\u65e6\u5927\u5b66\u6570\u5b66\u79d1\u5b66\u5b66\u9662\u5e94\u7528\u6570\u5b66\u65b9\u5411\u535a\u58eb\u7814\u7a76\u751f\u8d75\u4f2f\u6797\u3001\u5317\u4eac\u5e08\u8303\u5927\u5b66\u7cfb\u7edf\u79d1\u5b66\u5b66\u9662\u535a\u58eb\u7814\u7a76\u751f\u725f\u7267\u4e91\uff0c\u5171\u540c\u53d1\u8d77\u300c\u590d\u6742\u7cfb\u7edf\u81ea\u52a8\u5efa\u6a21\u300d\u8bfb\u4e66\u4f1a\u7b2c\u4e8c\u5b63<\/a>\u3002<\/span><\/span><\/p>\n


    <\/span><\/span><\/p>\n

    \u8bfb\u4e66\u4f1a\u5c06\u4e8e9\u67085\u65e5\u8d77\u6bcf\u5468\u56db\u665a\u4e0a20:00-22:00\u8fdb\u884c\uff0c\u63a2\u8ba8\u56db\u4e2a\u6838\u5fc3\u6a21\u5757\uff1a\u6570\u636e\u9a71\u52a8\u7684\u590d\u6742\u7cfb\u7edf\u5efa\u6a21\u3001\u590d\u6742\u7f51\u7edc\u7ed3\u6784\u63a8\u65ad\u3001\u5177\u6709\u53ef\u89e3\u91ca\u6027\u7684\u590d\u6742\u7cfb\u7edf\u63a8\u65ad\uff08\u52a8\u529b\u5b66+\u7f51\u7edc\u7ed3\u6784\uff09\u3001\u5e94\u7528-\u8d85\u6750\u6599\u8bbe\u8ba1\u548c\u57ce\u5e02\u7cfb\u7edf\uff0c\u901a\u8fc7\u91cd\u70b9\u8ba8\u8bba75\u7bc7\u7ecf\u5178\u3001\u524d\u6cbf\u7684\u91cd\u8981\u6587\u732e\uff0c\u4ece\u9ed1\u76d2\uff08\u6570\u636e\u9a71\u52a8\uff09\u5230\u767d\u76d2\uff08\u53ef\u89e3\u91ca\u6027\uff09\uff0c\u9010\u6b65\u6355\u6349\u7cfb\u7edf\u7684\u201c\u672c\u8d28\u201d\u89c4\u5f8b\uff0c\u5e2e\u52a9\u5927\u5bb6\u66f4\u597d\u7684\u8ba4\u8bc6\u3001\u7406\u89e3\u3001\u9884\u6d4b\u3001\u63a7\u5236\u3001\u8bbe\u8ba1\u590d\u6742\u7cfb\u7edf\uff0c\u4e3a\u76f8\u5173\u9886\u57df\u7684\u7814\u7a76\u548c\u5e94\u7528\u63d0\u4f9b\u6d1e\u89c1\u3002\u6b22\u8fce\u611f\u5174\u8da3\u7684\u670b\u53cb\u62a5\u540d\u53c2\u4e0e\uff01<\/span><\/span><\/p>\n


    <\/span><\/span><\/p>\n

    \"\u56fe\u7247\"<\/span><\/a><\/section>\n

    <\/span><\/span><\/section>\n
    \u8be6\u60c5\u8bf7\u89c1\uff1a<\/span><\/span><\/section>\n

    \u590d\u6742\u7cfb\u7edf\u81ea\u52a8\u5efa\u6a21\u8bfb\u4e66\u4f1a\uff1a\u4ece\u6570\u636e\u9a71\u52a8\u5230\u53ef\u89e3\u91ca\u6027\uff0c\u63a2\u7d22\u7cfb\u7edf\u5185\u5728\u89c4\u5f8b\uff5c\u5185\u964475\u7bc7\u9886\u57df\u5fc5\u8bfb\u6587\u732e<\/a><\/span><\/strong><\/p>\n


    <\/span><\/p>\n


    <\/span><\/h1>\n
    \u63a8\u8350\u9605\u8bfb<\/span><\/span><\/strong><\/strong><\/strong><\/section>\n
    <\/strong><\/section>\n
    1. \u81ea\u76f8\u4f3c\u52a8\u529b\u5b66\u7684\u6570\u636e\u9a71\u52a8\u5efa\u6a21\uff5c\u96c6\u667a\u7814\u7a76\u4e2d\u5fc3\u6210\u679c<\/span><\/a><\/span><\/strong>
    <\/span><\/strong><\/span><\/section>\n
    2. Chaos\uff1a\u6570\u636e\u9a71\u52a8\u7684\u590d\u6742\u7cfb\u7edf\u5efa\u6a21\u7279\u520a\u6982\u89c8<\/span><\/a><\/span>
    <\/span><\/strong><\/strong><\/section>\n
    3. \u6570\u636e\u9a71\u52a8\u7684\u81ea\u76f8\u4f3c\u6027\u53d1\u73b0\uff1a\u89e3\u9501\u590d\u6742\u7269\u7406\u73b0\u8c61\u4e2d\u7684\u6807\u5ea6\u5f8b<\/span><\/a><\/span>
    <\/span><\/strong><\/strong><\/section>\n
    <\/strong><\/section>\n
    4. <\/span><\/strong><\/span>\u6e38\u620f\u5316\u79d1\u7814\u2014\u2014\u8ba9\u6211\u4eec\u7a81\u7834\u5185\u5377\u3001\u5171\u7814\u6d8c\u73b0<\/a><\/span><\/strong><\/strong><\/span><\/section>\n
    <\/strong>5. <\/span><\/strong>\u63a2\u7d22\u8005\u8ba1\u5212 | \u96c6\u667a\u4ff1\u4e50\u90e82025\u5185\u5bb9\u56e2\u961f\u62db\u52df\uff08\u5168\u804c&\u517c\u804c\uff09<\/a><\/span><\/strong><\/span><\/section>\n

    6. \u52a0\u5165\u96c6\u667a\uff0c\u73a9\u8f6c\u590d\u6742\uff0c\u5171\u521b\u6591\u56fe\uff01\u96c6\u667a\u4ff1\u4e50\u90e8\u7ebf\u4e0b\u5fd7\u613f\u8005\u62db\u52df<\/span><\/a><\/span><\/strong>
    <\/span><\/strong><\/span><\/p>\n


    <\/span><\/p>\n


    <\/span><\/span><\/p>\n

    \u70b9\u51fb\u201c\u9605\u8bfb\u539f\u6587\u201d\uff0c\u62a5\u540d\u8bfb\u4e66\u4f1a<\/span><\/span><\/section>\n<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

    \u5bfc\u8bed \u5728\u5f53\u4eca\u5927\u6570\u636e\u65f6\u4ee3\uff0c\u6570\u636e\u5df2\u6210\u4e3a\u63a8\u52a8\u51b3\u7b56\u7684\u6838\u5fc3\u8d44\u6e90\u3002\u7136\u800c\uff0c\u5982\u4f55\u5c06\u6d77\u91cf\u6570\u636e\u8f6c\u5316\u4e3a\u6700\u4f18\u51b3\u7b56\uff0c\u4ecd\u662f\u5b66\u672f\u754c\u548c\u5de5\u4e1a\u754c\u9762\u4e34\u7684\u91cd\u5927\u6311\u6218\u3002\u8fd1\u65e5\u5317\u4eac\u5927\u5b66\u5b8b\u6d01\u6559\u6388\u56e2\u961f\u5728Cell Press\u4ea4\u53c9\u5b66\u79d1\u671f\u520aNexus\u53d1\u8868\u9898\u4e3a\u300aBridging Prediction and Decision: Advances and Ch…<\/p>\n","protected":false},"author":0,"featured_media":59889,"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\/59921"}],"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"}],"replies":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=59921"}],"version-history":[{"count":0,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/59921\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/media\/59889"}],"wp:attachment":[{"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=59921"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=59921"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=59921"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fspecial&post=59921"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}