{"id":5582,"date":"2024-10-07T03:22:00","date_gmt":"2024-10-06T18:22:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5582"},"modified":"2024-10-07T03:22:00","modified_gmt":"2024-10-06T18:22:00","slug":"improving-autonomous-ai-agents-with-reflective-tree-search-and-self-learning","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5582","title":{"rendered":"Improving Autonomous AI Agents with Reflective Tree Search and Self-Learning"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Improving Autonomous AI Agents with Reflective Tree Search and Self-Learning\u00a0<\/strong>[78.4]<br>Reflective Monte Carlo Tree Search (R-MCTS)\u306f\u3001AI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u80fd\u529b\u3092\u9ad8\u3081\u308b\u305f\u3081\u306b\u8a2d\u8a08\u3055\u308c\u305f\u65b0\u3057\u3044\u30c6\u30b9\u30c8\u30bf\u30a4\u30e0\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3067\u3042\u308b\u3002 R-MCTS\u306f1)\u5f93\u6765\u306eMCTS\u3092\u62e1\u5f35\u3057\u3001\u5bfe\u7167\u7684\u306a\u53cd\u5c04\u3092\u53d6\u308a\u5165\u308c\u3001\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306f\u904e\u53bb\u306e\u76f8\u4e92\u4f5c\u7528\u304b\u3089\u5b66\u3076\u3053\u3068\u304c\u3067\u304d\u308b\u3002 \u81ea\u5df1\u5b66\u7fd2\u306b\u3088\u308aGPT-4o\u3092\u5fae\u8abf\u6574\u3059\u308b\u3053\u3068\u3067\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u6027\u80fd\u3092\u5411\u4e0a\u3055\u305b\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2410.02052v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2410.02052v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 02 Oct 2024 21:42:35 GMT)<\/li>\n\n\n\n<li>\u300cWe propose Reflective Monte Carlo Tree Search (R-MCTS), an extension of classic MCTS that improves the agent\u2019s decision making process on the fly by incorporating reflection over its past task executions, and state estimations using multi-agent-debate\u300d\u3068\u3044\u3046\u30bf\u30a4\u30d7\u306e\u30e2\u30f3\u30c6\u30ab\u30eb\u30ed\u6728\u63a2\u7d22\u306e\u63d0\u6848\u3068\u3001\u305d\u308c\u306b\u3088\u308bSFT\u3067\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u7d50\u679c\u3092\u6539\u5584\u3002ToT\u3084\u5358\u7d14\u306aMCTS\u3088\u308a\u512a\u308c\u305f\u7d50\u679c\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/jasonyux\/RMCTS-self-learning\">jasonyux\/RMCTS-self-learning \u00b7 GitHub<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[42,390],"class_list":["post-5582","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-autonomous-agent","tag-synthetic-data"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5582","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=5582"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5582\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5582"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5582"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5582"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}