{"id":5721,"date":"2024-11-05T06:44:00","date_gmt":"2024-11-04T21:44:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5721"},"modified":"2024-11-05T06:44:00","modified_gmt":"2024-11-04T21:44:00","slug":"autokaggle-a-multi-agent-framework-for-autonomous-data-science-competitions","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5721","title":{"rendered":"AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitions\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitions\u00a0<\/strong>[47.7]<br>AutoKaggle\u306f\u3001\u30b3\u30fc\u30c9\u5b9f\u884c\u3068\u5358\u4f53\u30c6\u30b9\u30c8\u3092\u7d44\u307f\u5408\u308f\u305b\u305f\u53cd\u5fa9\u7684\u306a\u958b\u767a\u30d7\u30ed\u30bb\u30b9\u3092\u5b9f\u88c5\u3057\u3001\u30b3\u30fc\u30c9\u306e\u6b63\u3057\u3055\u3068\u30ed\u30b8\u30c3\u30af\u306e\u6574\u5408\u6027\u3092\u4fdd\u8a3c\u3059\u308b\u3002 \u30c7\u30fc\u30bf\u30af\u30ea\u30fc\u30cb\u30f3\u30b0\u3001\u7279\u5fb4\u5de5\u5b66\u3001\u30e2\u30c7\u30ea\u30f3\u30b0\u306e\u305f\u3081\u306e\u691c\u8a3c\u6e08\u307f\u6a5f\u80fd\u3092\u542b\u3080\u6c4e\u7528\u30c7\u30fc\u30bf\u30b5\u30a4\u30a8\u30f3\u30b9\u30c4\u30fc\u30eb\u30ad\u30c3\u30c8\u306f\u3001\u3053\u306e\u30bd\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u306e\u57fa\u790e\u3092\u5f62\u6210\u3057\u307e\u3059\u3002 AutoKaggle\u306f\u3001\u4e00\u822c\u7684\u306a\u30c7\u30fc\u30bf\u30b5\u30a4\u30a8\u30f3\u30b9\u30d1\u30a4\u30d7\u30e9\u30a4\u30f3\u306b\u304a\u3051\u308b\u30d0\u30ea\u30c7\u30fc\u30b7\u30e7\u30f3\u30ec\u30fc\u30c80.85\u3068\u7dcf\u5408\u30b9\u30b3\u30a20.82\u3092\u9054\u6210\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2410.20424v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2410.20424v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Sun, 27 Oct 2024 12:44:25 GMT)<\/li>\n\n\n\n<li>Kaggle\u306e\u3088\u3046\u306a\u30c7\u30fc\u30bf\u5206\u6790\u306e\u81ea\u52d5\u5316\u3002\u5bfe\u8c61\u3068\u3057\u3066\u3044\u308b\u30bf\u30b9\u30af\uff08\u5206\u6790\u30d5\u30a7\u30fc\u30ba\uff09\u306f\u300cbackground understanding, preliminary exploratory data analysis, data cleaning (DC), in-depth exploratory data analysis, feature engineering (FE), and model building, validation, and prediction (MBVP).\u300d\u3067\u901a\u5e38\u306eAutoML\u3088\u308a\u5e83\u3044\u3001\u5bfe\u8c61\u30c7\u30fc\u30bf\u306f\u30c6\u30fc\u30d6\u30eb\u30c7\u30fc\u30bf\u306e\u3088\u3046\u3002<\/li>\n\n\n\n<li>\u300cAs our analysis relies on GPT-4o, which is trained on data available until October 2023, it includes most of the Classic Kaggle competitions.To evaluate the generalization capabilities of AutoKaggle, we therefore focus on competitions initiated after 2024.\u300d\u3068Leak\u306b\u306f\u6c17\u3092\u4f7f\u3063\u3066\u3044\u308b\u3068\u306f\u3044\u3048\u3001\u300cEvaluation results demonstrate that AutoKaggle achieves a validation submission rate of 0.85 and a comprehensive score of 0.82 in typical data science pipelines, fully proving its effectiveness and practicality in handling complex data science tasks.\u300d\u3068\u3044\u3046\u8a00\u3044\u304d\u308a\u306f\u51c4\u3044\u3002\u3082\u3063\u3068\u3082\u3001\u4eca\u306eLLM\u306e\u6027\u80fd\u304b\u3089\u3057\u3066\u9069\u5207\u306a\u30d1\u30a4\u30d7\u30e9\u30a4\u30f3\u3092\u7d44\u3081\u3070\u89e3\u3051\u305d\u3046\u306a\u554f\u984c\u3067\u3042\u308b\u3068\u3044\u3046\u611f\u899a\u306f\u3042\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/multimodal-art-projection\/AutoKaggle\">GitHub &#8211; multimodal-art-projection\/AutoKaggle<\/a><\/li>\n<\/ul>\n\n\n\n<p><\/p>\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":[40,42,223,498],"class_list":["post-5721","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-automl","tag-autonomous-agent","tag-llm","tag-498"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5721","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=5721"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5721\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5721"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5721"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5721"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}