feed.xml

<?xml version='1.0' encoding='utf-8'?>
<rss version="2.0"><channel><title>Musicaldg</title><link>https://musicaldg.github.io</link><description>Musicaldg 的机器学习笔记。</description><language>zh-CN</language><item><title>IRL、RLHF 与 DPO</title><link>https://musicaldg.github.io/blog/irl-rlhf-dpo/</link><guid>https://musicaldg.github.io/blog/irl-rlhf-dpo/</guid><description>从逆强化学习出发，说明如何从专家行为与人类偏好学习目标，并连接到 RLHF 与 DPO。</description><pubDate>Tue, 03 Nov 2026 00:00:00 GMT</pubDate></item><item><title>Distribution Matching 与 GAIL</title><link>https://musicaldg.github.io/blog/distribution-matching-gail/</link><guid>https://musicaldg.github.io/blog/distribution-matching-gail/</guid><description>从轨迹与状态—动作分布匹配出发，说明 Forward KL 为何可能退化为 BC，以及 GAIL 如何用判别器完成分布匹配。</description><pubDate>Mon, 02 Nov 2026 00:00:00 GMT</pubDate></item><item><title>Behavior Cloning 与 PLID</title><link>https://musicaldg.github.io/blog/behavior-cloning-plid/</link><guid>https://musicaldg.github.io/blog/behavior-cloning-plid/</guid><description>说明 Behavior Cloning 的状态分布偏移，并推导 PLID / DAgger 如何用交互式专家标签修正策略。</description><pubDate>Tue, 20 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Active Learning</title><link>https://musicaldg.github.io/blog/active-learning/</link><guid>https://musicaldg.github.io/blog/active-learning/</guid><description>介绍主动学习的基本流程，比较不确定性采样、QBC 和 BALD 三种选样方法。</description><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate></item></channel></rss>