{"id":40094,"date":"2025-04-12T23:02:40","date_gmt":"2025-04-12T15:02:40","guid":{"rendered":"https:\/\/17aitech.com\/?p=40094"},"modified":"2025-04-12T23:02:40","modified_gmt":"2025-04-12T15:02:40","slug":"%e9%ad%94%e6%94%b9alphazero%e5%90%8e%ef%bc%8c%e3%80%8a%e6%88%91%e7%9a%84%e4%b8%96%e7%95%8c%e3%80%8bai%e8%80%81%e7%8e%a9%e5%ae%b6%e9%97%ae%e4%b8%96%ef%bc%8c%e5%b9%b2%e6%b4%bb%e4%b8%8d%e7%94%a8%e4%b8%8b","status":"publish","type":"post","link":"https:\/\/17aitech.com\/?p=40094","title":{"rendered":"\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4"},"content":{"rendered":"<p>\u6587\u7ae0\u6765\u6e90\u4e8e\u4e92\u8054\u7f51:<a href=\"https:\/\/www.jiqizhixin.com\/articles\/2025-04-12-5\" target=\"_blank\">\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4<\/a><\/p>\n<blockquote>\n<p>\u4eba\u548c\u667a\u80fd\u4f53\u5171\u4eab\u5956\u52b1\u53c2\u6570\uff0c\u8fd9\u624d\u662f\u5f3a\u5316\u5b66\u4e60\u6b63\u786e\u7684\u65b9\u5411\uff1f<\/p>\n<\/blockquote>\n<p>\u5927\u6a21\u578b\u9a71\u52a8\u7684 AI \u52a9\u624b\u53c8\u5347\u7ea7\u4e86\u3002\u672c\u5468\u4e94\uff0c\u79d1\u6280\u5708\u6b63\u5728\u56f4\u89c2\u4e00\u4e2a\u966a\u4f60\u4e00\u8d77\u73a9\u300a\u6211\u7684\u4e16\u754c\u300b\u7684 AI\u3002<\/p>\n<section><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif\"><\/a><\/section>\n<p>\u5b83\u8bdd\u4e0d\u591a\u8bf4\uff0c\u5c31\u662f\u57cb\u5934\u5e72\u6d3b\u3002\u4e00\u8d77\u76d6\u623f\u5b50\u7684\u65f6\u5019\uff0c\u4f60\u4e0d\u9700\u8981\u7ed9 AI \u4e00\u5f20\u84dd\u56fe\uff0c\u6216\u662f\u4e0d\u65ad\u544a\u8bc9\u5b83\u8be5\u600e\u4e48\u505a\uff0c\u4f60\u53ea\u9700\u8981\u76d6\u81ea\u5df1\u7684\uff0c\u5b83\u5c31\u80fd\u4e00\u8fb9\u89c2\u5bdf\u4e00\u904d\u914d\u5408\uff0c\u5e76\u89c2\u5bdf\u4f60\u7684\u610f\u56fe\u968f\u65f6\u6539\u53d8\u8ba1\u5212\u3002<\/p>\n<section><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-42c6f32265db93d30bb0303bbed19609.gif\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-42c6f32265db93d30bb0303bbed19609.gif\"><\/a><\/section>\n<p>\u73b0\u5728\uff0cAI \u53ef\u4ee5\u4e0d\u65ad\u4e3b\u52a8\u5b66\u4e60\u3001\u7ea0\u6b63\u9519\u8bef\uff0c\u5c55\u73b0\u51fa\u4e86\u6b64\u524d\u5927\u6a21\u578b\u667a\u80fd\u4f53\u65e0\u6cd5\u5b9e\u73b0\u7684\u4e00\u7cfb\u5217\u80fd\u529b\u3002<br \/><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-3dbd502d937236f6c7ef530e0f57ac81.gif\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-3dbd502d937236f6c7ef530e0f57ac81.gif\"><\/a><\/p>\n<p>\u770b\u8d77\u6765\uff0c\u65b0\u7248\u7684 AI \u5728\u4e0e\u6211\u4eec\u5171\u540c\u6e38\u620f\u65f6\u4e0d\u518d\u662f\u50ac\u4e00\u4e0b\u52a8\u4e00\u4e0b\u4e86\uff0c\u5b83\u5df2\u7ecf\u662f\u4e00\u4e2a\u6709\u300c\u4e3b\u89c2\u80fd\u52a8\u6027\u300d\u7684\u73a9\u5bb6\uff0c\u5c31\u50cf\u4e2a\u548c\u4f60\u5171\u540c\u73a9\u8fc7\u51e0\u767e\u5c40\u6e38\u620f\u7684\u8001\u53cb\u4e00\u6837\u3002<\/p>\n<p>\u8fd9\u9879\u6280\u672f\u540d\u4e3a AssistanceZero\uff0c\u51fa\u81ea\u52a0\u5dde\u5927\u5b66\u4f2f\u514b\u5229\u5206\u6821\uff08UC Berkeley\uff09\u3002\u503c\u5f97\u6ce8\u610f\u7684\u662f\uff0c\u5b83\u5e76\u672a\u63a5\u53d7\u5927\u6a21\u578b\u5e38\u89c1\u7684 RLHF \u8bad\u7ec3\u3002\u76f8\u53cd\uff0c\u5b83\u662f\u7531\u300cassistance games\u300d\u5f3a\u5316\u5b66\u4e60\u9a71\u52a8\u7684\uff0c\u7814\u7a76\u4eba\u5458\u8ba4\u4e3a\uff0c\u8fd9\u662f\u6784\u5efa AI \u52a9\u624b\u7684\u66f4\u597d\u9014\u5f84\u3002<\/p>\n<p>AI \u5728\u8fd9\u4e2a\u6846\u67b6\u4e2d\u5e76\u4e0d\u4f1a\u88ab\u52a8\u5730\u63a5\u53d7\u4eba\u7c7b\u53cd\u9988\uff0c\u800c\u662f\u5bfb\u6c42\u4e3b\u52a8\u4e0e\u4eba\u5408\u4f5c\uff0c\u901a\u8fc7\u63a8\u65ad\u76ee\u6807\u800c\u4e0d\u65ad\u4f18\u5316\u884c\u4e3a\uff0c\u8fd9\u907f\u514d\u4e86 RLHF \u4e2d AI \u53ef\u80fd\u4f1a\u51fa\u73b0\u7684\u4f5c\u5f0a\u884c\u4e3a\uff0c\u8ba9 AI \u53ef\u4ee5\u91c7\u53d6\u66f4\u52a0\u534f\u4f5c\u7684\u7b56\u7565\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-cb43a5860dbce7e53291e279f1b7c5c5.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-cb43a5860dbce7e53291e279f1b7c5c5.png\"><\/a><\/p>\n<ul>\n<li>\n<p>\u8bba\u6587\uff1aAssistanceZero: Scalably Solving Assistance Games<\/p>\n<\/li>\n<li>\n<p>\u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2504.07091<\/p>\n<\/li>\n<li>\n<p>\u9879\u76ee\u94fe\u63a5\uff1ahttps:\/\/github.com\/cassidylaidlaw\/minecraft-building-assistance-game<\/p>\n<\/li>\n<\/ul>\n<p>\u76ee\u6807\uff1a\u628a RLHF \u9769\u547d\u6389<\/p>\n<p>\u6700\u8fd1\uff0cAI \u9886\u57df\u91cc\u5f88\u591a\u7814\u7a76\u90fd\u5728\u5c1d\u8bd5\u6539\u8fdb\u6216\u66ff\u4ee3 RLHF\u3002<\/p>\n<p>\u6211\u4eec\u77e5\u9053\uff0c\u9884\u8bad\u7ec3\u3001\u76d1\u7763\u5fae\u8c03\uff08SFT\uff09\u4ee5\u53ca\u57fa\u4e8e\u4eba\u7c7b\u53cd\u9988\u7684\u5f3a\u5316\u5b66\u4e60\uff08RLHF\uff09\u6216\u5176\u53d8\u4f53\u5df2\u7ecf\u6210\u4e3a\u8bad\u7ec3\u901a\u7528 AI \u52a9\u624b\u7684\u4e3b\u8981\u8303\u5f0f\u3002RLHF \u6d89\u53ca\u5bf9\u9884\u8bad\u7ec3\u7684\u57fa\u7840\u6a21\u578b\u8fdb\u884c\u5fae\u8c03\uff0c\u4f7f\u5176\u6839\u636e\u4eba\u7c7b\u6807\u6ce8\u8005\u5bf9\u8bf8\u5982\u300c\u5e2e\u52a9\u6027\u300d\u548c\u300c\u65e0\u5bb3\u6027\u300d\u7b49\u6807\u51c6\u7684\u504f\u597d\u6765\u91c7\u53d6\u884c\u52a8\uff08\u5373\u751f\u6210\u54cd\u5e94\uff09\u3002\u7136\u800c\uff0c\u901a\u8fc7 RLHF \u8bad\u7ec3\u7684\u52a9\u624b\u5b58\u5728\u4e00\u4e9b\u7f3a\u70b9\uff1a<\/p>\n<p>1. \u6807\u6ce8\u8005\u53ef\u80fd\u4f1a\u88ab\u8bef\u5bfc\uff0c\u5bf9\u65e0\u5e2e\u52a9\u7684\u884c\u52a8\u7ed9\u4e88\u79ef\u6781\u7684\u53cd\u9988\uff0c\u4ece\u800c\u6fc0\u52b1\u52a9\u624b\u4ea7\u751f\u6b3a\u9a97\u6027\u6216\u64cd\u7eb5\u6027\u7684\u884c\u4e3a\u3002<\/p>\n<p>2.RLHF \u5e76\u4e0d\u9f13\u52b1\u6a21\u578b\u4fdd\u6301\u5bf9\u7528\u6237\u76ee\u6807\u7684\u4e0d\u786e\u5b9a\u6027\uff0c\u751f\u6210\u9ad8\u8bc4\u5206\u5355\u8f6e\u54cd\u5e94\u7684\u76ee\u6807\u4f7f\u5f97\u52a9\u624b\u4e0d\u613f\u63d0\u51fa\u6f84\u6e05\u95ee\u9898\u6216\u5bf9\u5176\u54cd\u5e94\u8fdb\u884c\u4fdd\u7559\uff0c\u50cf GitHub Copilot \u8fd9\u6837\u7684\u975e\u804a\u5929\u578b\u4eba\u5de5\u667a\u80fd\u52a9\u624b\u4e5f\u5b58\u5728\u7c7b\u4f3c\u7684\u95ee\u9898\uff0c\u5f53\u7f16\u7801\u4efb\u52a1\u4e0d\u660e\u786e\u65f6\uff0cCopilot \u65e0\u6cd5\u8981\u6c42\u6f84\u6e05\u3002<\/p>\n<p>3. \u50cf Copilot \u8fd9\u6837\u7684\u81ea\u52a8\u5b8c\u6210\u52a9\u624b\u4e5f\u5e76\u672a\u5145\u5206\u8003\u8651\u52a9\u624b\u884c\u4e3a\u7684\u534f\u4f5c\u6027 \u2014\u2014AI \u52a9\u624b\u7684\u884c\u52a8\u5e94\u8865\u5145\u7528\u6237\u7684\u884c\u4e3a\uff0c\u800c\u4e0d\u662f\u7b80\u5355\u5730\u9884\u6d4b\u6216\u53d6\u4ee3\u5b83\u4eec\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-1999e281f0c95ccc9e508b5d3368eec6.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-1999e281f0c95ccc9e508b5d3368eec6.png\"><\/a><\/p>\n<p>ChatGPT \u503e\u5411\u4e8e\u7528\u4e00\u4e2a\u56de\u590d\u6765\u89e3\u51b3\u4f60\u7684\u6240\u6709\u95ee\u9898\u3002\u5982\u679c\u4f60\u8981\u6c42 ChatGPT\u300c\u6e05\u7406\u4e00\u4e9b\u78c1\u76d8\u7a7a\u95f4\u300d\uff0c\u5b83\u4f1a\u7ed9\u4f60\u4e00\u4e2a\u7a0b\u5e8f\u8fd0\u884c\uff0c\u800c\u4e0d\u4f1a\u8be2\u95ee\u54ea\u4e9b\u6587\u4ef6\u53ef\u4ee5\u5220\u9664\u3002<\/p>\n<p>\u8bad\u7ec3 AI \u52a9\u624b\u7684\u53e6\u4e00\u79cd\u8303\u5f0f\u662f Assistance Games\u3002\u5b83\u901a\u8fc7\u660e\u786e\u8003\u8651\u534f\u52a9\u7684\u4e92\u52a8\u6027\u548c\u7528\u6237\u76ee\u6807\u7684\u4e0d\u786e\u5b9a\u6027\uff0c\u907f\u514d\u4e86 RLHF \u7684\u4e0a\u8ff0\u7f3a\u70b9\u3002<\/p>\n<p>\u5177\u4f53\u6765\u8bf4\uff0cAssistance Games \u662f\u4e00\u4e2a\u53cc\u4eba\u6e38\u620f\uff0c\u52a9\u624b\u548c\u7528\u6237\u5728\u4e00\u4e2a\u5171\u4eab\u73af\u5883\u4e2d\u91c7\u53d6\u884c\u52a8\uff08\u56fe 3b\uff09\u3002\u4e24\u4e2a Agent \u5171\u4eab\u4e00\u4e2a\u5956\u52b1\u51fd\u6570\uff0c\u4f46\u5173\u952e\u662f\u52a9\u624b\u6700\u521d\u5bf9\u8fd9\u4e2a\u5956\u52b1\u51fd\u6570\u662f\u4e0d\u786e\u5b9a\u7684\uff0cassistance games \u6d88\u9664\u4e86\u6b3a\u9a97\u7684\u52a8\u673a\uff0c\u56e0\u4e3a\u52a9\u624b\u7684\u8868\u73b0\u53d6\u51b3\u4e8e\u771f\u5b9e\u7684\u6f5c\u5728\u5956\u52b1\u51fd\u6570\uff0c\u800c\u4e0d\u662f\u4eba\u7c7b\u7684\u53cd\u9988\u3002\u6b64\u5916\uff0cassistance games \u8fd8\u6fc0\u52b1\u52a9\u624b\u4e0e\u7528\u6237\u4e92\u52a8\u4ee5\u89e3\u51b3\u5176\u4e0d\u786e\u5b9a\u6027\u3002\u6700\u540e\uff0c\u89e3\u51b3 assistance games \u7684\u7ed3\u679c\u662f\u52a9\u624b\u7684\u884c\u52a8\u80fd\u591f\u8865\u5145\u7528\u6237\u7684\u884c\u52a8\uff0c\u4ee5\u5b9e\u73b0\u6700\u4f18\u7684\u8054\u5408\u6027\u80fd\u3002\u800c\u4e14\uff0c\u7814\u7a76\u4eba\u5458\u8fd8\u8bbe\u60f3\u4e86\u4e00\u79cd\u5c06 assistance games \u5e94\u7528\u4e8e\u5927\u8bed\u8a00\u6a21\u578b\u540e\u8bad\u7ec3\u7684\u65b9\u6cd5\uff0c\u4ee5\u66ff\u4ee3 RLHF\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-f93d05ead2747556a7638c9d248fc943.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-f93d05ead2747556a7638c9d248fc943.png\"><\/a><\/p>\n<p>\u5c3d\u7ba1 Assistance Games \u5177\u6709\u8bf8\u591a\u4f18\u52bf\uff0c\u4f46\u5b83\u4eec\u4e3a\u4f55\u4ecd\u7136\u662f\u4e00\u4e2a\u7814\u7a76\u8f83\u5c11\u7684\u8bad\u7ec3 AI \u52a9\u624b\u7684\u65b9\u6cd5\u5462\uff1fAssistance Games \u8fc7\u53bb\u4ec5\u88ab\u7528\u4e8e\u89e3\u51b3\u975e\u5e38\u7b80\u5355\u7684\u95ee\u9898\uff0c\u4f46\u5728\u590d\u6742\u73af\u5883\u4e2d\u5374\u88ab\u5e7f\u6cdb\u5ffd\u89c6\uff0c\u4e3b\u8981\u662f\u7531\u4e8e\u4ee5\u4e0b\u770b\u4f3c\u4e0d\u53ef\u514b\u670d\u7684\u6311\u6218\uff1a<\/p>\n<p>1. \u8ba1\u7b97\u4e0a\u7684\u96be\u9898\uff1aAI \u52a9\u624b\u9700\u8981\u5728\u5956\u52b1\u51fd\u6570\u7684\u4e0d\u786e\u5b9a\u6027\u4e0b\u4fdd\u6301\u51b3\u7b56\u80fd\u529b\uff0c\u800c\u8fd9\u88ab\u8ba4\u4e3a\u662f\u8ba1\u7b97\u4e0a\u4e0d\u53ef\u884c\u3002<\/p>\n<p>2. \u4eba\u7c7b\u6a21\u578b\u7684\u51c6\u786e\u6027\uff1a\u4e0e RLHF \u4e0d\u540c\uff0c\u89e3\u51b3 Assistance Games \u9700\u8981\u4e00\u4e2a\u80fd\u591f\u51c6\u786e\u9884\u6d4b\u4eba\u7c7b\u5bf9 AI \u884c\u52a8\u53cd\u5e94\u7684\u6a21\u578b\u3002\u5982\u679c AI \u65e0\u6cd5\u7406\u89e3\u4eba\u7c7b\u7684\u6c9f\u901a\u7b56\u7565\uff0c\u53ef\u80fd\u4f1a\u5728\u4e0e\u771f\u5b9e\u4eba\u7c7b\u4e92\u52a8\u65f6\u8868\u73b0\u4e0d\u4f73\u3002\u8fc7\u53bb\u5173\u4e8e Assistance Games \u7684\u7814\u7a76\u4f7f\u7528\u4e86\u57fa\u4e8e\u5f3a\u5316\u5b66\u4e60\u6216\u89c4\u5212\u7684\u4eba\u7c7b\u6a21\u578b\uff0c\u4f46\u8fd9\u4e9b\u6a21\u578b\u53ef\u80fd\u4e0e\u771f\u5b9e\u4eba\u7c7b\u884c\u4e3a\u6709\u663e\u8457\u5dee\u5f02\u3002<\/p>\n<p>\u8be5\u7814\u7a76\u56e2\u961f\u6210\u529f\u5e94\u5bf9\u4e86\u8fd9\u4e9b\u6311\u6218\uff0c\u5e76\u8bc1\u660e\u4e86\u590d\u6742\u7684 Assistance Games \u662f\u53ef\u4ee5\u88ab\u6709\u6548\u89e3\u51b3\u7684\u3002\u4e3a\u6b64\uff0c\u4ed6\u4eec\u5f15\u5165\u4e86\u4e00\u4e2a\u65b0\u57fa\u51c6\u6d4b\u8bd5 \u2014\u2014Minecraft Building Assistance Game (MBAG)\u3002<\/p>\n<p>\u5728\u8fd9\u4e2a\u6d4b\u8bd5\u4e2d\uff0cAI \u52a9\u624b\u9700\u8981\u5728\u300a\u6211\u7684\u4e16\u754c\u300b\u6e38\u620f\u73af\u5883\u4e2d\u5e2e\u52a9\u4eba\u7c7b\u5efa\u9020\u76ee\u6807\u7ed3\u6784\uff0c\u4f46\u52a9\u624b\u5bf9\u76ee\u6807\u4e00\u65e0\u6240\u77e5\uff08\u56fe 1\uff09\u3002MBAG \u7684\u6311\u6218\u5728\u4e8e\u76ee\u6807\u7ed3\u6784\u7684\u5206\u5e03\u975e\u5e38\u590d\u6742\uff0c\u53ef\u80fd\u7684\u76ee\u6807\u6570\u91cf\u8d85\u8fc7 10^400 \u4e2a\uff0c\u8fdc\u8fdc\u8d85\u8fc7\u4ee5\u5f80\u7814\u7a76\u4e2d\u7684\u6570\u91cf\uff0c\u540c\u65f6\u72b6\u6001\u548c\u52a8\u4f5c\u7a7a\u95f4\u4e5f\u66f4\u5927\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-54aac4b99b5a502ff5655396be926207.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-54aac4b99b5a502ff5655396be926207.png\"><\/a><\/p>\n<p>\u7814\u7a76\u4eba\u5458\u901a\u8fc7 MBAG \u7814\u7a76\u4e86\u6df1\u5ea6\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u662f\u5426\u80fd\u591f\u89e3\u51b3 Assistance Games\u3002\u7814\u7a76\u53d1\u73b0\uff0cPPO\uff08\u4e00\u79cd\u6d41\u884c\u7684\u65e0\u6a21\u578b\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff09\u53ef\u4ee5\u8f7b\u677e\u5730\u5728 MBAG \u4e2d\u5efa\u9020\u5df2\u77e5\u76ee\u6807\u623f\u5c4b\uff0c\u4f46\u5728\u76ee\u6807\u7ed3\u6784\u672a\u77e5\u65f6\u8868\u73b0\u4e0d\u4f73\u3002\u4ed6\u4eec\u8ba4\u4e3a\u8fd9\u662f\u56e0\u4e3a PPO \u9700\u8981\u540c\u65f6\u4ece\u9ad8\u65b9\u5dee\u7684\u53cd\u9988\u4e2d\u5b66\u4e60\u5982\u4f55\u9884\u6d4b\u76ee\u6807\u5e76\u6839\u636e\u9884\u6d4b\u91c7\u53d6\u884c\u52a8\uff0c\u8fd9\u589e\u52a0\u4e86\u5176\u96be\u5ea6\u3002<\/p>\n<p>\u56e0\u6b64\uff0c\u4e3a\u66f4\u597d\u5730\u89e3\u51b3 Assistance Games \u95ee\u9898\uff0c\u4ed6\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u540d\u4e3a AssistanceZero \u7684\u65b0\u7b97\u6cd5\uff0c\u8be5\u7b97\u6cd5\u901a\u8fc7\u6269\u5c55 AlphaZero \u6765\u5206\u79bb\u9884\u6d4b\u548c\u884c\u52a8\u3002\u4e0e AlphaZero \u7c7b\u4f3c\uff0cAssistanceZero \u7ed3\u5408\u4e86\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\uff08MCTS\uff09\u548c\u795e\u7ecf\u7f51\u7edc\u6765\u9009\u62e9\u884c\u52a8\u3002AssistanceZero \u91c7\u7528\u4e86\u4e00\u79cd\u5177\u6709\u989d\u5916\u8f93\u51fa\u5c42\u7684\u795e\u7ecf\u7f51\u7edc\uff0c\u8fd9\u4e9b\u8f93\u51fa\u5c42\u7528\u4e8e\u9884\u6d4b\u5956\u52b1\u548c\u4eba\u7c7b\u884c\u4e3a\uff0cMCTS \u5229\u7528\u8fd9\u4e9b\u9884\u6d4b\u5728\u4e0d\u786e\u5b9a\u6027\u4e0b\u6709\u6548\u89c4\u5212\uff08\u56fe 4\uff09\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-5395258bc58763cc6b22360e6553738e.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-5395258bc58763cc6b22360e6553738e.png\"><\/a><\/p>\n<p>AssistanceZero \u7684\u6548\u679c\u8fdc\u80dc\u4e8e PPO\uff08\u8868 1\uff09\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-66cb83d66724f4703e59442b70ad3630.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-66cb83d66724f4703e59442b70ad3630.png\"><\/a><\/p>\n<p>\u6b64\u5916\uff0c\u4ed6\u4eec\u8fd8\u901a\u8fc7\u63a2\u7d22\u5982\u4f55\u5f00\u53d1\u51fa\u80fd\u4ea7\u751f\u6709\u6548\u52a9\u624b\u7684\u4eba\u7c7b\u6a21\u578b\u6765\u5e94\u5bf9\u89e3\u51b3 Assistance Games \u7684\u7b2c\u4e8c\u4e2a\u6311\u6218\u3002\u6709\u8da3\u7684\u662f\uff0c\u4ed6\u4eec\u53d1\u73b0 MBAG \u4e2d\u6700\u4f73\u7684\u4eba\u7c7b\u6a21\u578b\u4e5f\u7ed3\u5408\u4e86 MCTS \u548c\u6a21\u4eff\u5b66\u4e60\uff0c\u8fd9\u79cd\u65b9\u6cd5\u88ab\u79f0\u4e3a piKL\u3002<\/p>\n<p>\u7814\u7a76\u4eba\u5458\u5c06\u901a\u8fc7 Assistance Games \u8bad\u7ec3\u7684\u7b56\u7565\u4e0e\u5176\u4ed6\u65b9\u6cd5\uff08\u5982\u7c7b\u4f3c\u9884\u8bad\u7ec3\u548c SFT \u7684\u6d41\u7a0b\uff09\u8bad\u7ec3\u7684\u7b56\u7565\u8fdb\u884c\u4e86\u6bd4\u8f83\u3002<\/p>\n<p>\u5728 MBAG \u4e2d\uff0c\u4ed6\u4eec\u53d1\u73b0\u901a\u8fc7 AssistanceZero \u8bad\u7ec3\u7684\u52a9\u624b\u5728\u6700\u4f73\u4eba\u7c7b\u6a21\u578b\uff08\u8868 3\uff09\u548c\u771f\u5b9e\u4eba\u7c7b\uff08\u56fe 2\uff09\u65b9\u9762\u90fd\u5927\u5927\u4f18\u4e8e\u901a\u8fc7\u9884\u8bad\u7ec3 + SFT \u6216\u5176\u4ed6\u65b9\u6cd5\u8bad\u7ec3\u7684\u52a9\u624b\u3002AssistanceZero \u52a9\u624b\u5c55\u73b0\u4e86\u8bb8\u591a\u6709\u7528\u7684\u81ea\u53d1\u884c\u4e3a\uff0c\u4f8b\u5982\u6839\u636e\u4fee\u6b63\u8fdb\u884c\u9002\u5e94\uff08\u56fe 1\uff09\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-2efe8b77e15fd45a5ea7d44e9493cd27.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-2efe8b77e15fd45a5ea7d44e9493cd27.png\"><\/a><\/p>\n<p><sup>\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u88683<\/sup><\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-3c0af89367123b92854a5d5db86db91e.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-3c0af89367123b92854a5d5db86db91e.png\"><\/a><\/p>\n<p><sup>\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u56fe2<\/sup><\/p>\n<p>\u603b\u7684\u6765\u8bf4\uff0c\u7ed3\u679c\u8868\u660e\uff0cAssistance Games \u662f\u53ef\u6269\u5c55\u7684\uff0c\u5e76\u4e14\u53ef\u4ee5\u6210\u4e3a\u5728\u5177\u6709\u6311\u6218\u6027\u7684\u73af\u5883\u4e2d\u8bad\u7ec3\u6709\u7528\u52a9\u624b\u7684\u4f18\u8d8a\u6846\u67b6\u3002<\/p>\n<p>\u4ec0\u4e48\u662fMBAG<\/p>\n<p>\u5728\u8bbe\u8ba1 MBAG \u65f6\uff0c\u7814\u7a76\u4eba\u5458\u8bbe\u5b9a\u4e86\u51e0\u4e2a\u76ee\u6807\uff0c\u4ee5\u4f7f\u5176\u6210\u4e3a\u4e00\u4e2a\u66f4\u5e7f\u6cdb\u7814\u7a76\u534f\u52a9\u6e38\u620f\u7684\u6709\u7528\u73af\u5883\u3002<\/p>\n<p>\u5176\u8bbe\u8ba1\u76ee\u6807\u5305\u62ec\u590d\u6742\u7684\u5956\u52b1\u53c2\u6570\u5206\u5e03\u3001\u591a\u6837\u7684\u52a9\u624b\u5e2e\u52a9\u65b9\u5f0f\uff0c\u4ee5\u53ca\u9002\u5408\u5b66\u672f\u5b9e\u9a8c\u5ba4\u8bad\u7ec3 RL agent \u7684\u73af\u5883\u3002<\/p>\n<p>MBAG \u662f\u7531\u4e00\u4e2a\u4e09\u7ef4\u65b9\u5757\u7f51\u683c\u3001\u7f51\u683c\u5185\u7684\u73a9\u5bb6\u4f4d\u7f6e\u4ee5\u53ca\u73a9\u5bb6\u7684\u7269\u54c1\u680f\u7ec4\u6210\u3002\u7f51\u683c\u4e2d\u7684\u6bcf\u4e2a\u4f4d\u7f6e\u53ef\u4ee5\u662f\u5341\u79cd\u65b9\u5757\u7c7b\u578b\u4e4b\u4e00\uff0c\u5305\u62ec\u7a7a\u6c14\uff0c\u5b9e\u9a8c\u4e2d\u4f7f\u7528\u4e86\u4e00\u4e2a 11\u00d710\u00d710 \u7684\u7f51\u683c\u3002<\/p>\n<p>\u52a8\u4f5c\u7a7a\u95f4\u5305\u62ec\u65e0\u64cd\u4f5c\u3001\u79fb\u52a8\u3001\u653e\u7f6e\u65b9\u5757\u548c\u7834\u574f\u65b9\u5757\u3002\u653e\u7f6e\u548c\u7834\u574f\u52a8\u4f5c\u7531\u4f4d\u7f6e\u53c2\u6570\u5316\uff0c\u653e\u7f6e\u52a8\u4f5c\u8fd8\u7531\u65b9\u5757\u7c7b\u578b\u53c2\u6570\u5316\uff0c\u8fd9\u610f\u5473\u7740\u5728 11\u00d710\u00d710 \u7684\u73af\u5883\u4e2d\uff0c\u6709\u8d85\u8fc7 20000 \u79cd\u53ef\u80fd\u7684\u52a8\u4f5c\u3002<\/p>\n<p>\u73a9\u5bb6\u53ea\u80fd\u5230\u8fbe\u6709\u9650\u7684\u8ddd\u79bb\u6765\u7834\u574f\u6216\u653e\u7f6e\u65b9\u5757\uff0c\u800c\u4e14\u5728\u5f53\u524d\u72b6\u6001\u4e0b\uff0c\u8bb8\u591a\u52a8\u4f5c\u662f\u65e0\u6548\u7684\uff08\u4f8b\u5982\uff0c\u4e0d\u53ef\u80fd\u7834\u574f\u7a7a\u6c14\u65b9\u5757\uff09\u3002\u56e0\u6b64\uff0c\u901a\u5e38\u53ea\u6709\u4e00\u5c0f\u90e8\u5206\u52a8\u4f5c\u662f\u6709\u6548\u7684\u3002<\/p>\n<p>\u63d0\u51fa AssistanceZero \u65b0\u7b97\u6cd5<\/p>\n<p>\u7814\u7a76\u4eba\u5458\u4f7f\u7528 MBAG \u6765\u7814\u7a76\u5982\u4f55\u89e3\u51b3\u534f\u52a9\u6e38\u620f\u4e2d\u7684\u590d\u6742\u5e8f\u8d2f\u51b3\u7b56\u95ee\u9898\uff0c\u5e76\u5c1d\u8bd5\u4e86 PPO\uff08\u4e00\u79cd\u65e0\u6a21\u578b\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\uff09\u8bad\u7ec3\u52a9\u624b\u7b56\u7565\u3002<\/p>\n<p>\u7136\u800c\uff0c\u4ed6\u4eec\u53d1\u73b0 PPO \u5728 MBAG \u4e2d\u8868\u73b0\u4e0d\u4f73\u3002\u4f7f\u7528\u5faa\u73af PPO \u8bad\u7ec3\u7684\u52a9\u624b\u6839\u672c\u65e0\u6cd5\u5e2e\u52a9\u4eba\u7c7b\u6a21\u578b\uff08\u89c1\u8868 1 \u7684\u7b2c\u4e00\u884c\uff09\uff0c\u800c\u975e\u5faa\u73af PPO \u7684\u8868\u73b0\u7565\u4f18\u4e8e\u5faa\u73af PPO\uff08\u89c1\u7b2c\u4e8c\u884c\uff09\u3002\u4ed6\u4eec\u8ba4\u4e3a\uff0c\u8fd9\u662f\u56e0\u4e3a\u9ad8\u65b9\u5dee\u7684\u5956\u52b1\u4fe1\u53f7\u4f7f\u5f97 PPO \u96be\u4ee5\u6709\u6548\u5b66\u4e60\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-5013b55c430e6fd9c7d6ae0dd2b651fb.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-5013b55c430e6fd9c7d6ae0dd2b651fb.png\"><\/a><\/p>\n<p><sup>\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u88681<\/sup><\/p>\n<p>\u6b64\u5916\uff0c\u7531\u4e8e\u52a9\u624b\u5bf9\u76ee\u6807\u7ed3\u6784\u4e0d\u786e\u5b9a\uff0c\u5373\u4f7f\u6839\u636e\u89c2\u5bdf\u5386\u53f2\uff0c\u91c7\u53d6\u4e00\u4e2a\u5728\u671f\u671b\u4e2d\u6709\u5e2e\u52a9\u7684\u52a8\u4f5c\u6709\u65f6\u4e5f\u4f1a\u5bfc\u81f4\u8d1f\u9762\u5956\u52b1\u3002\u4efb\u52a1\u7684\u5e8f\u8d2f\u6027\u548c\u957f\u671f\u6027\u52a0\u5267\u4e86\u8fd9\u4e9b\u95ee\u9898\uff0c\u8fdb\u4e00\u6b65\u589e\u52a0\u4e86 PPO \u8bd5\u56fe\u4f18\u5316\u7684\u5956\u52b1\u4fe1\u53f7\u7684\u566a\u58f0\u3002<\/p>\n<p>\u5728\u8bad\u7ec3\u521d\u671f\uff0cPPO \u63a5\u6536\u5230\u7684\u6700\u660e\u663e\u7684\u4fe1\u53f7\u662f\u653e\u7f6e\u548c\u7834\u574f\u52a8\u4f5c\u5f80\u5f80\u662f\u9519\u8bef\u7684\uff0c\u5bfc\u81f4\u8d1f\u9762\u5956\u52b1\u3002\u56e0\u6b64\uff0c\u52a9\u624b\u7b56\u7565\u6536\u655b\u5230\u51e0\u4e4e\u4e0d\u5efa\u9020\u4efb\u4f55\u4e1c\u897f\u3002\u4e3a\u4e86\u51cf\u5c11\u5956\u52b1\u4fe1\u53f7\u4e2d\u7684\u566a\u58f0\uff0c\u5e76\u6fc0\u52b1\u52a9\u624b\u66f4\u591a\u5730\u91c7\u53d6\u884c\u52a8\uff0c\u4ed6\u4eec\u63a2\u7d22\u4e86\u4ec5\u6839\u636e\u52a9\u624b\u81ea\u8eab\u52a8\u4f5c\u7684\u5956\u52b1\u6765\u8bad\u7ec3\u52a9\u624b\uff0c\u8fd8\u5c1d\u8bd5\u6dfb\u52a0\u4e00\u4e2a\u8f85\u52a9\u635f\u5931\u9879\uff0c\u4ee5\u9f13\u52b1\u653e\u7f6e\u6b63\u786e\u7684\u65b9\u5757\u3002<\/p>\n<p>\u8fd9\u4e9b\u65b9\u6cd5\u7565\u5fae\u63d0\u9ad8\u4e86\u52a9\u624b &#8211; \u4eba\u7c7b\u6a21\u578b\u7ec4\u5408\u5b8c\u6210\u7684\u76ee\u6807\u767e\u5206\u6bd4\uff0c\u540c\u65f6\u51cf\u5c11\u4e86\u4eba\u7c7b\u6a21\u578b\u7684\u52a8\u4f5c\u6570\u91cf\u6216\u4fdd\u6301\u5176\u4e0d\u53d8\uff08\u89c1\u8868 1 \u7684\u7b2c\u4e09\u548c\u7b2c\u56db\u884c\uff09\u3002\u7136\u800c\uff0c\u5b83\u4eec\u4ecd\u7136\u53ea\u662f\u52c9\u5f3a\u6709\u5e2e\u52a9\u3002<\/p>\n<p>\u4e3a\u4e86\u89e3\u51b3 PPO \u7684\u5c40\u9650\u6027\uff0c\u4ed6\u4eec\u8bbe\u8ba1\u4e86 AssistanceZero \u6765\u5206\u79bb\u76ee\u6807\u9884\u6d4b\u548c\u884c\u52a8\u9009\u62e9\uff0c\u901a\u8fc7\u5b66\u4e60\u4e00\u4e2a\u76ee\u6807\u9884\u6d4b\u5668\uff0c\u7136\u540e\u5c06\u5176\u7528\u4e8e\u89c4\u5212\u3002<\/p>\n<p>\u5177\u4f53\u6765\u8bf4\uff0cAssistanceZero \u662f AlphaZero \u7684\u6269\u5c55\uff0cAlphaZero \u662f\u4e00\u79cd\u5728\u56f4\u68cb\u548c\u56fd\u9645\u8c61\u68cb\u7b49\u590d\u6742\u7ade\u4e89\u6027\u6e38\u620f\u4e2d\u53d6\u5f97\u8d85\u4eba\u8868\u73b0\u7684\u6df1\u5ea6\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u3002<\/p>\n<p>\u548c AlphaZero \u4e00\u6837\uff0cAssistanceZero \u4f7f\u7528\u8499\u7279\u5361\u6d1b\u6811\u641c\u7d22\uff08MCTS\uff09\u7684\u4e00\u4e2a\u53d8\u4f53\u6765\u9009\u62e9\u52a8\u4f5c\u3002MCTS \u901a\u8fc7\u6a21\u62df\u4ece\u5f53\u524d\u72b6\u6001\u91c7\u53d6\u4e0d\u540c\u52a8\u4f5c\u5e8f\u5217\u7684\u7ed3\u679c\u6765\u6784\u5efa\u641c\u7d22\u6811\u3002\u7136\u800c\uff0c\u5b83\u9700\u8981\u77e5\u9053\u5956\u52b1\u548c\u52a8\u4f5c\u5bfc\u81f4\u7684\u4e0b\u4e00\u4e2a\u72b6\u6001\uff0c\u8fd9\u4e24\u8005\u5728\u534f\u52a9 POMDP \u4e2d\u90fd\u662f\u672a\u77e5\u7684\u3002<\/p>\n<p>\u867d\u7136\u4f5c\u8005\u5df2\u7ecf\u8bc1\u660e AssistanceZero \u53ef\u4ee5\u8bad\u7ec3\u51fa\u4e0e\u56fa\u5b9a\u4eba\u7c7b\u6a21\u578b\u914d\u5408\u826f\u597d\u7684\u52a9\u624b\uff0c\u4f46\u5982\u4f55\u5148\u83b7\u5f97\u4e00\u4e2a\u597d\u7684\u4eba\u7c7b\u6a21\u578b\u4ecd\u7136\u4e0d\u6e05\u695a\u3002\u7406\u60f3\u60c5\u51b5\u4e0b\uff0c\u52a9\u624b\u7b56\u7565\u4e0d\u4ec5\u5e94\u8be5\u4e0e\u8bad\u7ec3\u65f6\u4f7f\u7528\u7684\u4eba\u7c7b\u6a21\u578b\u8868\u73b0\u826f\u597d\uff0c\u8fd8\u5e94\u8be5\u4e0e\u771f\u5b9e\u4eba\u7c7b\u914d\u5408\u65f6\u8868\u73b0\u826f\u597d\u3002\u4ed6\u4eec\u63a2\u7d22\u4e86\u4eba\u7c7b AI \u4ea4\u4e92\u6587\u732e\u4e2d\u5f00\u53d1 MBAG \u4eba\u7c7b\u6a21\u578b\u7684\u51e0\u79cd\u65b9\u6cd5\uff0c\u5305\u62ec\u57fa\u4e8e\u5956\u52b1\u548c\u57fa\u4e8e\u6570\u636e\u7684\u6a21\u578b\u3002<\/p>\n<p>\u57fa\u4e8e\u5956\u52b1\u7684\u4eba\u7c7b\u6a21\u578b\u5047\u8bbe\u4eba\u7c7b\u9009\u62e9\u52a8\u4f5c\u8fd1\u4f3c\u4e8e\u6700\u4f18\uff0c\u4ee5\u6700\u5927\u5316\u5176\u5956\u52b1\u51fd\u6570\u3002\u4ed6\u4eec\u4f7f\u7528\u6df1\u5ea6\u5f3a\u5316\u5b66\u4e60\u8bad\u7ec3\u4e86\u4e24\u4e2a\u57fa\u4e8e\u5956\u52b1\u7684\u6a21\u578b\u6765\u72ec\u81ea\u5efa\u9020\u76ee\u6807\u7ed3\u6784\u3002<\/p>\n<p>\u5bf9\u4e8e\u5176\u4e2d\u4e00\u4e2a\u6a21\u578b\uff0c\u4ed6\u4eec\u4f7f\u7528\u4e86\u5e26\u6709\u71b5\u7cfb\u6570\u7684 PPO\uff0c\u8fd1\u4f3c\u4e8e Boltzmann \u7406\u6027\uff0c\u8fd9\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u4eba\u7c7b\u884c\u4e3a\u7684\u566a\u58f0\u6700\u4f18\u6a21\u578b\uff1b\u53e6\u4e00\u4e2a\u6a21\u578b\u5219\u4f7f\u7528\u4e86 AlphaZero \u8bad\u7ec3\u3002<\/p>\n<p>\u63a5\u4e0b\u6765\uff0c\u4ed6\u4eec\u4f7f\u7528\u884c\u4e3a\u514b\u9686\uff08BC\uff09\u8bad\u7ec3\u4e86\u4e00\u7cfb\u5217\u57fa\u4e8e\u6570\u636e\u7684\u4eba\u7c7b\u6a21\u578b\uff0c\u884c\u4e3a\u514b\u9686\u4f7f\u7528\u76d1\u7763\u5b66\u4e60\u4ece\u72b6\u6001\u9884\u6d4b\u52a8\u4f5c\u3002\u5bf9\u4e8e\u8bad\u7ec3\u6570\u636e\u96c6\uff0c\u4ed6\u4eec\u8bb0\u5f55\u4e86\u4e94\u4e2a\u53d7\u8bd5\u8005\u5728 MBAG \u4e2d\u5efa\u9020\u623f\u5c4b\u7684 18 \u4e2a\u7247\u6bb5\u3002<\/p>\n<p>\u5728\u4e00\u534a\u7684\u7247\u6bb5\u4e2d\u4eba\u7c7b\u72ec\u81ea\u5efa\u9020\uff0c\u53e6\u4e00\u534a\u5219\u7531\u4e00\u4f4d\u6709\u7ecf\u9a8c\u7684\u300a\u6211\u7684\u4e16\u754c\u300b\u73a9\u5bb6\u4f5c\u4e3a\u52a9\u624b\u3002\u4ed6\u4eec\u5c06\u76ee\u6807\u7ed3\u6784\u663e\u793a\u4e3a\u53d7\u8bd5\u8005\u7684\u4e00\u4e2a\u534a\u900f\u660e\u84dd\u56fe\uff0c\u8986\u76d6\u5728\u6b63\u5e38\u7684\u6e38\u620f\u4e0a\uff0c\u540c\u65f6\u5bf9\u4eba\u7c7b\u52a9\u624b\u9690\u85cf\u76ee\u6807\u7ed3\u6784\u3002\u4f7f\u7528 BC\uff0c\u4ed6\u4eec\u8bad\u7ec3\u4e86\u4e09\u79cd\u4eba\u7c7b\u6a21\u578b\uff1a\u4e00\u79cd\u57fa\u4e8e\u53d7\u8bd5\u8005\u72ec\u81ea\u6e38\u620f\u7684\u6570\u636e\uff08BC-alone\uff09\uff0c\u4e00\u79cd\u57fa\u4e8e\u4e0e\u52a9\u624b\u4e00\u8d77\u6e38\u620f\u7684\u5b50\u96c6\uff08BC-with-assistant\uff09\uff0c\u4ee5\u53ca\u4e00\u79cd\u57fa\u4e8e\u6574\u4e2a\u6570\u636e\u96c6\uff08BC-combined\uff09\u3002<\/p>\n<p>\u867d\u7136\u7814\u7a76\u4eba\u5458\u5bf9 Assistance Games \u7684\u6b63\u5f0f\u5b9a\u4e49\u5047\u8bbe\u4eba\u7c7b\u6a21\u578b\u662f\u9a6c\u5c14\u53ef\u592b\u7684\uff0c\u4f46\u4ed6\u4eec\u53d1\u73b0\u57fa\u4e8e\u5faa\u73af\u3001\u5386\u53f2\u7684 BC \u6a21\u578b\u6bd4\u9a6c\u5c14\u53ef\u592b\u7b56\u7565\u66f4\u80fd\u9884\u6d4b\u4eba\u7c7b\u52a8\u4f5c\u3002\u9664\u4e86\u6355\u6349\u4e2a\u522b\u975e\u9a6c\u5c14\u53ef\u592b\u884c\u4e3a\u5916\uff0c\u5faa\u73af\u4eba\u7c7b\u6a21\u578b\u8fd8\u53ef\u4ee5\u9690\u5f0f\u5730\u6a21\u62df\u591a\u79cd\u4eba\u7c7b\u7b56\u7565\u7684\u6df7\u5408\u3002\u8fd9\u4f7f\u5f97\u4e00\u4e2a\u5355\u4e00\u7684\u5faa\u73af\u6a21\u578b\u6709\u53ef\u80fd\u6355\u6349\u5230\u771f\u5b9e\u4eba\u7c7b\u6280\u80fd\u6c34\u5e73\u7684\u5dee\u5f02\u3002<\/p>\n<p>\u6709\u671b\u63d0\u5347\u5927\u6a21\u578b\u540e\u8bad\u7ec3<\/p>\n<p>\u5728\u8be5\u7814\u7a76\u4e2d\uff0c\u7814\u7a76\u4eba\u5458\u5b9e\u73b0\u4e86\u901a\u8fc7 Assistance Games \u5728 MBAG \u4e2d\u8bad\u7ec3\u52a9\u624b\u7684\u5b8c\u6574\u65b9\u6848\uff0c\u7136\u540e\u4f7f\u7528 AssistanceZero \u6c42\u89e3\u751f\u6210\u7684\u8f85\u52a9 POMDP\u3002\u5982\u679c\u5c06 Assistance Games \u4e0e\u5176\u4ed6 AI \u52a9\u624b\u8bad\u7ec3\u8303\u5f0f\u8fdb\u884c\u6bd4\u8f83\u4f1a\u5982\u4f55\uff1f<\/p>\n<p>\u5177\u4f53\u800c\u8a00\uff0c\u4f5c\u8005\u5f00\u53d1\u4e86\u7528\u4e8e\u8bad\u7ec3 MBAG \u52a9\u624b\u7684\u6d41\u7a0b\uff0c\u7c7b\u4f3c\u4e8e GitHub Copilot\/OpenAI Codex \u548c RLHF \u7684\u76d1\u7763\u5fae\u8c03 (SFT) \u9636\u6bb5\u6240\u4f7f\u7528\u7684\u6d41\u7a0b\uff0c\u8fd9\u4e5f\u662f\u5f53\u524d\u8bad\u7ec3 AI \u52a9\u624b\u7684\u4e24\u4e2a\u4e3b\u8981\u8303\u5f0f\u3002\u7814\u7a76\u4eba\u5458\u5c06\u751f\u6210\u7684\u7b56\u7565\u4e0e\u4f7f\u7528 AssistanceZero \u8bad\u7ec3\u7684\u52a9\u624b\u8fdb\u884c\u6bd4\u8f83\u3002<\/p>\n<p>RLHF \u548c Codex \u90fd\u4ee5\u9884\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u4e3a\u8d77\u70b9\uff0c\u8fd9\u4f7f\u5f97\u5b83\u4eec\u80fd\u591f\u5b66\u4e60\u6709\u7528\u7684\u8868\u5f81\u5e76\u9884\u6d4b\u4eba\u7c7b\u884c\u4e3a\u3002\u5728 MBAG \u4e2d\uff0c\u4f5c\u8005\u751f\u6210\u4e86\u4e00\u4e2a\u9884\u8bad\u7ec3\u8bed\u6599\u5e93\uff0c\u4f7f\u7528\u7ed3\u5408 BC \u7684\u4eba\u7c7b\u6a21\u578b\u751f\u6210 1 \u4e07\u4e2a\u56de\u5408\uff0c\u8be5\u56de\u5408\u4f1a\u4ece\u8bad\u7ec3\u96c6 Dtrain \u4e2d\u968f\u673a\u9009\u62e9\u76ee\u6807\u7ed3\u6784\u8fdb\u884c\u6784\u5efa\u3002\u7136\u540e\u4ece\u89c2\u6d4b\u503c\u4e2d\u5220\u9664\u76ee\u6807\u7ed3\u6784\u4fe1\u606f\uff0c\u5e76\u5728\u751f\u6210\u7684\u6570\u636e\u96c6\u4e0a\u8bad\u7ec3\u4e00\u4e2a\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\uff08\u5373\u9884\u8bad\u7ec3\u6a21\u578b\uff09\u3002\u4e0e\u8bed\u8a00\u6216\u4ee3\u7801\u6a21\u578b\u7c7b\u4f3c\uff0c\u8be5\u6a21\u578b\u53ef\u4ee5\u5728\u6ca1\u6709\u76ee\u6807\u4fe1\u606f\u7684\u60c5\u51b5\u4e0b\u9884\u6d4b\u4eba\u7c7b\u884c\u4e3a\uff0c\u5e76\u5b66\u4e60\u5230\u4e86\u80fd\u591f\u7406\u89e3\u4eba\u7c7b\u76ee\u6807\u7ed3\u6784\u7684\u8868\u5f81\u3002<\/p>\n<p>\u901a\u8fc7\u5728\u4f4e\u70ed\u72b6\u6001\u4e0b\u4ece\u9884\u8bad\u7ec3\u6a21\u578b\u4e2d\u91c7\u6837\u52a8\u4f5c\uff0c\u6211\u4eec\u5c31\u83b7\u5f97\u4e86\u4e00\u4e2a\u7c7b\u4f3c\u4e8e GitHub Copilot \u7684\u52a9\u624b\uff1a\u5f53\u5b83\u5bf9\u4eba\u7c7b\u5c06\u91c7\u53d6\u7684\u52a8\u4f5c\u9ad8\u5ea6\u81ea\u4fe1\u65f6\uff0c\u5b83\u4f1a\u6784\u5efa\u76ee\u6807\u7ed3\u6784\uff0c\u7f3a\u4e4f\u81ea\u4fe1\u65f6\u5219\u4e0d\u4f1a\u91c7\u53d6\u884c\u52a8\u3002<\/p>\n<p>\u8fdb\u4e00\u6b65\u4f7f\u7528 SFT \u8bad\u7ec3\u9884\u8bad\u7ec3\u6a21\u578b\uff0c\u5176\u4e2d\u4f7f\u7528\u4eba\u7c7b\u4e13\u5bb6\u4f5c\u4e3a\u52a9\u624b\u7684\u6570\u636e\uff0c\u5bf9\u9884\u8bad\u7ec3\u6a21\u578b\u8fdb\u884c\u5fae\u8c03\u4ee5\u6a21\u4eff\u4eba\u7c7b\u52a9\u624b\uff0c\u7c7b\u4f3c\u4e8e\u5728 RLHF \u7684 SFT \u9636\u6bb5\u8bad\u7ec3 LLM \u6a21\u4eff\u4eba\u7c7b\u4e66\u5199\u7684\u52a9\u624b\u56de\u590d\u7684\u65b9\u5f0f\u3002\u7814\u7a76\u4f7f\u7528\u7f51\u683c\u641c\u7d22\u5bf9 540 \u4e2a\u8d85\u53c2\u6570\u7ec4\u5408\u8fdb\u884c\u641c\u7d22\uff0c\u4ee5\u627e\u5230\u9002\u5408 SFT \u7b56\u7565\u7684\u5b66\u4e60\u7387\u3001\u8bad\u7ec3\u5468\u671f\u3001\u6570\u636e\u589e\u5f3a\u548c dropout \u7684\u6700\u4f73\u7ec4\u5408\u3002<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-88f63c6740a4452d7e4b24a02101254d.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-88f63c6740a4452d7e4b24a02101254d.png\"><\/a><\/p>\n<p>\u8868 3 \u6bd4\u8f83\u4e86\u9884\u8bad\u7ec3\u6a21\u578b\u548c SFT \u6a21\u578b\u4ee5\u53ca\u57fa\u4e8e Assistance Games \u7684\u7b56\u7565\u3002\u4f5c\u8005\u4f7f\u7528\u7ed3\u5408 piKL \u7684\u4eba\u7c7b\u6a21\u578b\u5bf9\u6bcf\u4e2a\u6a21\u578b\u8fdb\u884c\u4e86\u8d85\u8fc7 1000 \u8f6e\u8bc4\u4f30\uff0c\u5e76\u62a5\u544a\u4e86\u4e0e\u8868 1 \u76f8\u540c\u7684\u6307\u6807\u3002\u9884\u8bad\u7ec3\u7b56\u7565\u548c SFT \u7b56\u7565\u5747\u7565\u5fae\u51cf\u5c11\u4e86\u5b9e\u73b0\u76f8\u4f3c\u76ee\u6807\u5b8c\u6210\u7387\u6240\u9700\u7684\u4eba\u7c7b\u64cd\u4f5c\u6570\u91cf\uff08\u7ea6 4-5 \u4e2a\uff09\u3002SFT \u7b56\u7565\u5e73\u5747\u6784\u5efa\u4e86\u7ea6 3% \u7684\u76ee\u6807\u7ed3\u6784\u3002\u76f8\u6bd4\u4e4b\u4e0b\uff0c\u4f7f\u7528 AssistanceZero \u8bad\u7ec3\u7684\u7b56\u7565\u5c06\u4eba\u7c7b\u64cd\u4f5c\u6570\u91cf\u51cf\u5c11\u4e86\u7ea6 65 \u4e2a\uff0c\u540c\u65f6\u63d0\u9ad8\u4e86\u76ee\u6807\u5b8c\u6210\u7387\uff1b\u5b83\u6784\u5efa\u4e86\u7ea6 26% \u7684\u76ee\u6807\u3002<\/p>\n<p>\u4f5c\u8005\u8fd8\u6bd4\u8f83\u4e86 AI \u52a9\u624b\u4e0e\u771f\u4eba\u7684\u8868\u73b0\u3002\u6bd4\u8f83\u56db\u79cd\u6761\u4ef6\u4e0b\u7684\u4eba\u7c7b\u73a9\u5bb6\uff1a\u72ec\u81ea\u4e00\u4eba\uff08\u65e0\u52a9\u624b\uff09\u3001\u4f7f\u7528 SFT \u7b56\u7565\u3001\u4f7f\u7528 AssistanceZero \u8bad\u7ec3\u7684\u52a9\u624b\u4ee5\u53ca\u4e0e\u4e13\u5bb6\u4eba\u7c7b\u52a9\u624b\u4e00\u8d77\u8fdb\u884c\u6e38\u620f\uff0c\u6bcf\u4e2a\u53c2\u4e0e\u8005\u8fde\u7eed\u4e94\u6b21\u5efa\u9020\u540c\u4e00\u680b\u623f\u5c4b\u3002\u7b2c\u4e00\u6b21\u7528\u4e8e\u7ec3\u4e60\uff0c\u5e2e\u52a9\u53d7\u8bd5\u8005\u719f\u6089\u300a\u6211\u7684\u4e16\u754c\u300b\u7684\u64cd\u4f5c\u548c\u76ee\u6807\u7ed3\u6784\uff0c\u968f\u540e\u53d7\u8bd5\u8005\u5728\u56db\u79cd\u6761\u4ef6\u4e0b\u4ee5\u968f\u673a\u987a\u5e8f\u5efa\u9020\u623f\u5c4b\u3002<\/p>\n<p>\u5728\u6bcf\u6b21\u4e92\u52a8\u7ed3\u675f\u540e\uff0c\u53d7\u8bd5\u8005\u5bf9\u5176\u6574\u4f53\u5b9e\u7528\u6027\u8fdb\u884c\u8bc4\u5206\uff0c\u7ed3\u679c\u663e\u793a\u7ecf\u8fc7 AssistanceZero \u8bad\u7ec3\u7684\u52a9\u624b\u8868\u73b0\u660e\u663e\u4f18\u4e8e SFT \u52a9\u624b\uff0c\u5e76\u63a5\u8fd1\u4eba\u7c7b\u57fa\u51c6\u3002\u5176\u4e2d\uff0c\u53c2\u4e0e\u8005\u5bf9 AssistanceZero \u80fd\u591f\u4ece\u7ea0\u6b63\u4e2d\u6709\u6548\u5b66\u4e60\u7684\u80fd\u529b\u5370\u8c61\u6df1\u523b\u3002\u4f8b\u5982\uff0c\u5728\u4eba\u7c7b\u7834\u574f\u4e00\u4e24\u4e2a\u9519\u8bef\u65b9\u5757\u540e\uff0cAssistantZero \u4e5f\u80fd\u7834\u574f\u591a\u4e2a\u9519\u8bef\u65b9\u5757\uff0c\u76f8\u6bd4\u4e4b\u4e0b SFT \u52a9\u624b\u5219\u5b8c\u5168\u6ca1\u6709\u5e2e\u52a9\u3002<\/p>\n<p>\u4f2f\u514b\u5229\u7684\u7814\u7a76\u4eba\u5458\u5e0c\u671b\uff0c\u57fa\u4e8e Assistance Games \u7684\u5de5\u4f5c\u6700\u7ec8\u53ef\u4ee5\u5e2e\u52a9\u5927\u8bed\u8a00\u6a21\u578b\u5b9e\u73b0\u89e3\u51b3\u590d\u6742\u95ee\u9898\u7684\u80fd\u529b\u3002<\/p>\n<p><sup>\u53c2\u8003\u5185\u5bb9\uff1a<\/sup><\/p>\n<p><sup>https:\/\/x.com\/cassidy_laidlaw\/status\/1910708807258534008<\/sup><\/p>\n<p>\u6587\u7ae0\u6765\u6e90\u4e8e\u4e92\u8054\u7f51:<a href=\"https:\/\/www.jiqizhixin.com\/articles\/2025-04-12-5\" target=\"_blank\">\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u6587\u7ae0\u6765\u6e90\u4e8e\u4e92\u8054\u7f51:\u9b54\u6539AlphaZero [&hellip;]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"open","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[27],"tags":[70,68],"class_list":["post-40094","post","type-post","status-publish","format-standard","hentry","category-news","tag-agent","tag-68"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4 - \u4e00\u8d77AI\u6280\u672f<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/17aitech.com\/?p=40094\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/17aitech.com\/?p=40094\",\"url\":\"https:\/\/17aitech.com\/?p=40094\",\"name\":\"\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4 - \u4e00\u8d77AI\u6280\u672f\",\"isPartOf\":{\"@id\":\"https:\/\/17aitech.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/17aitech.com\/?p=40094#primaryimage\"},\"image\":{\"@id\":\"https:\/\/17aitech.com\/?p=40094#primaryimage\"},\"thumbnailUrl\":\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif\",\"datePublished\":\"2025-04-12T15:02:40+00:00\",\"author\":{\"@id\":\"https:\/\/17aitech.com\/#\/schema\/person\/60225458499e817ae0af73e67e440b9d\"},\"breadcrumb\":{\"@id\":\"https:\/\/17aitech.com\/?p=40094#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/17aitech.com\/?p=40094\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/17aitech.com\/?p=40094#primaryimage\",\"url\":\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif\",\"contentUrl\":\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/17aitech.com\/?p=40094#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/17aitech.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/17aitech.com\/#website\",\"url\":\"https:\/\/17aitech.com\/\",\"name\":\"\u4e00\u8d77AI\u6280\u672f\",\"description\":\"\u8ba9AI\u77e5\u8bc6\u89e6\u624b\u53ef\u53ca\",\"alternateName\":\"\u4e00\u8d77AI\u6280\u672f\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/17aitech.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/17aitech.com\/#\/schema\/person\/60225458499e817ae0af73e67e440b9d\",\"name\":\"AI\u5c0f\u52a9\u624b\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/17aitech.com\/#\/schema\/person\/image\/\",\"url\":\"\/\/17aitech.com\/wp-content\/uploads\/2024\/04\/robot_3.png\",\"contentUrl\":\"\/\/17aitech.com\/wp-content\/uploads\/2024\/04\/robot_3.png\",\"caption\":\"AI\u5c0f\u52a9\u624b\"},\"description\":\"\u8fd9\u4e2a\u4eba\u5f88\u61d2\uff0c\u4ec0\u4e48\u90fd\u6ca1\u6709\u7559\u4e0b\uff5e\",\"url\":\"https:\/\/17aitech.com\/?page_id=33738&user=3\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4 - \u4e00\u8d77AI\u6280\u672f","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/17aitech.com\/?p=40094","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/17aitech.com\/?p=40094","url":"https:\/\/17aitech.com\/?p=40094","name":"\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4 - \u4e00\u8d77AI\u6280\u672f","isPartOf":{"@id":"https:\/\/17aitech.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/17aitech.com\/?p=40094#primaryimage"},"image":{"@id":"https:\/\/17aitech.com\/?p=40094#primaryimage"},"thumbnailUrl":"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif","datePublished":"2025-04-12T15:02:40+00:00","author":{"@id":"https:\/\/17aitech.com\/#\/schema\/person\/60225458499e817ae0af73e67e440b9d"},"breadcrumb":{"@id":"https:\/\/17aitech.com\/?p=40094#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/17aitech.com\/?p=40094"]}]},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/17aitech.com\/?p=40094#primaryimage","url":"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif","contentUrl":"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/04\/frc-612f46a8c5ec7f25938f09660831e652.gif"},{"@type":"BreadcrumbList","@id":"https:\/\/17aitech.com\/?p=40094#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/17aitech.com\/"},{"@type":"ListItem","position":2,"name":"\u9b54\u6539AlphaZero\u540e\uff0c\u300a\u6211\u7684\u4e16\u754c\u300bAI\u8001\u73a9\u5bb6\u95ee\u4e16\uff0c\u5e72\u6d3b\u4e0d\u7528\u4e0b\u6307\u4ee4"}]},{"@type":"WebSite","@id":"https:\/\/17aitech.com\/#website","url":"https:\/\/17aitech.com\/","name":"\u4e00\u8d77AI\u6280\u672f","description":"\u8ba9AI\u77e5\u8bc6\u89e6\u624b\u53ef\u53ca","alternateName":"\u4e00\u8d77AI\u6280\u672f","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/17aitech.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/17aitech.com\/#\/schema\/person\/60225458499e817ae0af73e67e440b9d","name":"AI\u5c0f\u52a9\u624b","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/17aitech.com\/#\/schema\/person\/image\/","url":"\/\/17aitech.com\/wp-content\/uploads\/2024\/04\/robot_3.png","contentUrl":"\/\/17aitech.com\/wp-content\/uploads\/2024\/04\/robot_3.png","caption":"AI\u5c0f\u52a9\u624b"},"description":"\u8fd9\u4e2a\u4eba\u5f88\u61d2\uff0c\u4ec0\u4e48\u90fd\u6ca1\u6709\u7559\u4e0b\uff5e","url":"https:\/\/17aitech.com\/?page_id=33738&user=3"}]}},"_links":{"self":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts\/40094","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=40094"}],"version-history":[{"count":0,"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts\/40094\/revisions"}],"wp:attachment":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=40094"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=40094"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=40094"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}