<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>交叉熵 on BrightPan Blog</title><link>https://www.brightpan.app/tags/%E4%BA%A4%E5%8F%89%E7%86%B5/</link><description>Recent content in 交叉熵 on BrightPan Blog</description><generator>Hugo -- 0.146.0</generator><language>en-us</language><lastBuildDate>Thu, 16 Apr 2026 19:00:00 +0800</lastBuildDate><atom:link href="https://www.brightpan.app/tags/%E4%BA%A4%E5%8F%89%E7%86%B5/index.xml" rel="self" type="application/rss+xml"/><item><title>硬币 vs 违约：从一个对比走通信息论与概率统计</title><link>https://www.brightpan.app/programming/coin_vs_default_information_theory_and_statistics/</link><pubDate>Thu, 16 Apr 2026 19:00:00 +0800</pubDate><guid>https://www.brightpan.app/programming/coin_vs_default_information_theory_and_statistics/</guid><description>抛硬币和预测违约，都是二元结果、都有概率——但为什么一个不需要模型，另一个养活了整个行业？沿着这个对比，一层层展开信息论与概率统计的核心概念。</description></item><item><title>信息论学习笔记：从数学基础到工程应用，再到语言现象</title><link>https://www.brightpan.app/programming/information_theory_learning_notes/</link><pubDate>Thu, 16 Apr 2026 14:45:00 +0800</pubDate><guid>https://www.brightpan.app/programming/information_theory_learning_notes/</guid><description>一份偏复习型的信息论学习笔记：先把熵、条件熵、联合熵、互信息和交叉熵讲清楚，再看它们在计算机、通信、风控和 LLM 中怎么用，最后发散到诗词、文言文与日常语言的“信息密度”。</description></item></channel></rss>