<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[L74211的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/L74211</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; L74211]]></copyright><item><title><![CDATA[强化学习算法演进与 DeepSeek-V4 后训练方法论（不抠公式，做概念讲解）]]></title><link>https://blog.youkuaiyun.com/L74211/article/details/163635478</link><guid>https://blog.youkuaiyun.com/L74211/article/details/163635478</guid><author>L74211</author><pubDate>Mon, 10 Aug 2026 12:56:49 +0800</pubDate><description><![CDATA[本文从强化学习（Reinforcement Learning, RL）的算法谱系出发，以 Value-based 与 Policy-based 两大分支为框架，系统梳理了从 Monte Carlo、Q-learning 到 GRPO 的算法演进脉络，着重阐明各节点之间的继承关系与设计动机。]]></description><category></category></item></channel></rss>