https://txtmix.com/tags/rl/Recent content in RL on Text MatrixHugozh-cnSat, 23 May 2026 08:55:34 +0800https://txtmix.com/posts/tech/mathematical-foundations-of-reinforcement-learning-book/Fri, 17 Apr 2026 16:05:00 +0800https://txtmix.com/posts/tech/mathematical-foundations-of-reinforcement-learning-book/<h1 id="mathematical-foundations-of-reinforcement-learning强化学习的数学基石">Mathematical Foundations of Reinforcement Learning：强化学习的数学基石</h1> <blockquote> <p><strong>目标读者</strong>：计算机科学/人工智能研究生、RL研究者、工程师 <strong>前置知识</strong>：概率论、线性代数基础 <strong>特色</strong>：网格世界（Grid World）贯穿全书的统一示例，数学严谨但叙述友好 <strong>难度定位</strong>：⭐⭐⭐⭐ 专家设计</p>https://txtmix.com/posts/tech/ai-agent/agent-lightning-microsoft-ai-agent-training-guide/Wed, 01 Apr 2026 01:20:00 +0800https://txtmix.com/posts/tech/ai-agent/agent-lightning-microsoft-ai-agent-training-guide/<h1 id="agent-lightning微软-ai-智能体强化学习训练框架完全指南">Agent Lightning：微软 AI 智能体强化学习训练框架完全指南</h1> <blockquote> <p>预计阅读时间：35分钟 | 难度：⭐⭐⭐⭐</p></blockquote> <hr> <h2 id="2-项目概述">§2 项目概述</h2> <h3 id="21-什么是-agent-lightning">2.1 什么是 Agent Lightning？</h3> <p><strong>Agent Lightning</strong>（<a href="https://github.com/microsoft/agent-lightning" target="_blank" rel="noopener noreffer ">GitHub 仓库</a>）是微软研究院开发的 AI 智能体训练框架，核心理念是「The absolute trainer to light up AI agents」。</p>