https://blog.youkuaiyun.com/gitblog_07834zh-cnhttps://blog.youkuaiyun.com/https://blog.youkuaiyun.com/gitblog_07834/article/details/148942897https://blog.youkuaiyun.com/gitblog_07834/article/details/148942897gitblog_07834Mon, 21 Sep 2026 21:17:30 +0800 **导读**：本文基于蘑菇书 Easy-RL 的 [第 10 章 稀疏奖励](https://link.gitcode.com/i/4d9d5377565d6f372ef7067bf5487f90) 展开，系统剖析强化学习中最棘手的问题之一——奖励稀疏时智能体"什么都学不到"的困境，并完整讲解三类应对思路：设计奖励（Reward Shaping）、好奇心驱动的内在奖励（ICM）、课程学习（Cu]]>https://blog.youkuaiyun.com/gitblog_07834/article/details/151608437https://blog.youkuaiyun.com/gitblog_07834/article/details/151608437gitblog_07834Sun, 20 Sep 2026 11:17:12 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148730898https://blog.youkuaiyun.com/gitblog_07834/article/details/148730898gitblog_07834Wed, 20 May 2026 09:08:00 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148943118https://blog.youkuaiyun.com/gitblog_07834/article/details/148943118gitblog_07834Mon, 11 May 2026 14:45:46 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148969856https://blog.youkuaiyun.com/gitblog_07834/article/details/148969856gitblog_07834Sun, 10 May 2026 07:50:42 +0800Ascend 950PR/Ascend 950DT ]]>https://blog.youkuaiyun.com/gitblog_07834/article/details/148886005https://blog.youkuaiyun.com/gitblog_07834/article/details/148886005gitblog_07834Sun, 03 May 2026 12:33:36 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148523064https://blog.youkuaiyun.com/gitblog_07834/article/details/148523064gitblog_07834Tue, 21 Apr 2026 11:24:39 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148039449https://blog.youkuaiyun.com/gitblog_07834/article/details/148039449gitblog_07834Mon, 26 Jan 2026 17:15:21 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/151607807https://blog.youkuaiyun.com/gitblog_07834/article/details/151607807gitblog_07834Sun, 21 Dec 2025 02:08:43 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148506598https://blog.youkuaiyun.com/gitblog_07834/article/details/148506598gitblog_07834Wed, 03 Dec 2025 01:12:21 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/142229226https://blog.youkuaiyun.com/gitblog_07834/article/details/142229226gitblog_07834Sun, 16 Nov 2025 13:02:02 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148730718https://blog.youkuaiyun.com/gitblog_07834/article/details/148730718gitblog_07834Fri, 14 Nov 2025 09:44:15 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/151611942https://blog.youkuaiyun.com/gitblog_07834/article/details/151611942gitblog_07834Fri, 12 Sep 2025 10:30:40 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/149011139https://blog.youkuaiyun.com/gitblog_07834/article/details/149011139gitblog_07834Mon, 30 Jun 2025 09:03:41 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148969460https://blog.youkuaiyun.com/gitblog_07834/article/details/148969460gitblog_07834Sat, 28 Jun 2025 09:10:34 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148968687https://blog.youkuaiyun.com/gitblog_07834/article/details/148968687gitblog_07834Sat, 28 Jun 2025 09:08:57 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148943913https://blog.youkuaiyun.com/gitblog_07834/article/details/148943913gitblog_07834Fri, 27 Jun 2025 09:11:01 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148806234https://blog.youkuaiyun.com/gitblog_07834/article/details/148806234gitblog_07834Sat, 21 Jun 2025 09:58:18 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148806005https://blog.youkuaiyun.com/gitblog_07834/article/details/148806005gitblog_07834Sat, 21 Jun 2025 09:53:50 +0800https://blog.youkuaiyun.com/gitblog_07834/article/details/148805910https://blog.youkuaiyun.com/gitblog_07834/article/details/148805910gitblog_07834Sat, 21 Jun 2025 09:52:10 +0800