{"author_url":"https://blog.hatena.ne.jp/Gasyou/","blog_title":"GA\u5c06\uff1f\u958b\u767a\u65e5\u8a18\uff5e\u738b\u7406\u306e\u305d\u306e\u5148\u3078\uff5e","blog_url":"https://gasyou.hatenablog.jp/","published":"2010-10-14 19:52:36","image_url":null,"description":"\u4f8b\u306b\u3088\u3063\u3066\u81ea\u5206\u7528\u3067\u3059\u3002 \u30a2\u30d6\u30b9\u30c8\u30e9\u30af\u30c8 \u5927\u898f\u6a21\u306a\u5f37\u5316\u5b66\u7fd2\u554f\u984c\u3092\u89e3\u304f\u306b\u306f\u95a2\u6570\u8fd1\u4f3c\u304c\u5fc5\u8981\u3060\u3002\u904e\u53bb10\u5e74\u3067\u4e3b\u6d41\u3060\u3063\u305f\u306e\u306f\u4fa1\u5024\u95a2\u6570\u306e\u8fd1\u4f3c\uff08\u3068\u3001\u305d\u308c\u3092\u30d9\u30fc\u30b9\u306b\u3057\u305f\u65b9\u7b56\uff09\u3060\u3063\u305f\u3002\u3053\u308c\u306f\u591a\u304f\u306e\u5fdc\u7528\u306b\u304a\u3044\u3066\u826f\u304f\u50cd\u3044\u305f\u304c\u3001\u305d\u306e\u624b\u6cd5\u306b\u306f\u3044\u304f\u3064\u304b\u306e\u9650\u754c\u3082\u3042\u308b\u3002 \u3053\u306e\u8ad6\u6587\u3067\u306f\u5225\u306e\u95a2\u6570\u8fd1\u4f3c\u306e\u624b\u6cd5\u3092\u793a\u3059\u3002\u305d\u308c\u306f\u3001\uff08\u305d\u306e\u5185\u90e8\u306b\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u6301\u3064\uff09\u72ec\u7acb\u3057\u305f\u95a2\u6570\u8fd1\u4f3c\u5668\u3092\u7528\u3044\u3066\u3001\u78ba\u7387\u7684\u306a\u65b9\u7b56\u3092\u76f4\u63a5\u8a18\u8ff0\u3059\u308b\u624b\u6cd5\u3060\u3002Williams\u306eREINFORCE\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306f\u4fa1\u5024\u95a2\u6570\u3092\u7528\u3044\u308b\u5f37\u5316\u5b66\u7fd2\u624b\u6cd5\u3088\u308a\uff08\u53ce\u675f\u304c\uff1f\uff09\u9045\u3044\u3002\u4fa1\u5024\u95a2\u6570\u3092\u5b66\u7fd2\u3057\u3001\u52fe\u914d\u306e\u30d0\u30ea\u30a2\u30f3\u30b9\u3092\u4e88\u6e2c\u3059\u308b\u306e\u306b\u4f7f\u3046\u306e\u306f\u9ad8\u901f\u306a\u5b66\u7fd2\u306e\u70ba\u306b\u306f\u52b9\u679c\u7684\u3060\u3002 1. \u65b9\u7b56\u52fe\u914d\u5b9a\u7406 \u95a2\u6570\u8fd1\u4f3c\u3092\u7528\u3044\u308b\u2026","author_name":"Gasyou","provider_url":"https://hatena.blog","provider_name":"Hatena Blog","height":"190","version":"1.0","width":"100%","type":"rich","url":"https://gasyou.hatenablog.jp/entry/20101014/1287053556","categories":["\u958b\u767a\u65e5\u8a18","\u30e1\u30e2"],"title":"Policy Gradient Methods for Reinforcement Learning with Function Approximation\u8aad\u3093\u3060\u307e\u3068\u3081","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fgasyou.hatenablog.jp%2Fentry%2F20101014%2F1287053556\" title=\"Policy Gradient Methods for Reinforcement Learning with Function Approximation\u8aad\u3093\u3060\u307e\u3068\u3081 - GA\u5c06\uff1f\u958b\u767a\u65e5\u8a18\uff5e\u738b\u7406\u306e\u305d\u306e\u5148\u3078\uff5e\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>"}