<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Warmup on Chen Kai Blog</title><link>https://www.chenk.top/zh/tags/warmup/</link><description>Recent content in Warmup on Chen Kai Blog</description><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Sun, 18 Sep 2022 09:00:00 +0000</lastBuildDate><atom:link href="https://www.chenk.top/zh/tags/warmup/index.xml" rel="self" type="application/rss+xml"/><item><title>优化理论（四）：学习率与调度策略</title><link>https://www.chenk.top/zh/optimization-theory/04-learning-rate-schedules/</link><pubDate>Sun, 18 Sep 2022 09:00:00 +0000</pubDate><guid>https://www.chenk.top/zh/optimization-theory/04-learning-rate-schedules/</guid><description>&lt;p>模型崩溃了，你把学习率减半——模型终于能训练了，但速度慢得惊人；再减半，损失几乎不再下降，曲线趋于平缓。这种场景是不是很熟？在所有可调的超参数里，&lt;strong>学习率（learning rate, LR）是最容易决定训练成败的那一个&lt;/strong>——它直接决定了模型是顺利收敛、进展极其缓慢，还是迅速发散。&lt;/p></description></item></channel></rss>