<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Embedding on Chen Kai Blog</title><link>https://www.chenk.top/en/tags/embedding/</link><description>Recent content in Embedding on Chen Kai Blog</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sat, 13 Dec 2025 09:00:00 +0000</lastBuildDate><atom:link href="https://www.chenk.top/en/tags/embedding/index.xml" rel="self" type="application/rss+xml"/><item><title>Recommendation Systems (5): Embedding and Representation Learning</title><link>https://www.chenk.top/en/recommendation-systems/05-embedding-techniques/</link><pubDate>Sat, 13 Dec 2025 09:00:00 +0000</pubDate><guid>https://www.chenk.top/en/recommendation-systems/05-embedding-techniques/</guid><description>&lt;p>When Netflix suggests &lt;em>Inception&lt;/em> to someone who just finished &lt;em>The Dark Knight&lt;/em>, the magic is not a hand-crafted &amp;ldquo;if-watched-Nolan-then&amp;rdquo; rule. It is geometry. Both films sit close together in a 128-dimensional &lt;strong>embedding space&lt;/strong> that the model has learned from billions of viewing events. Geometry replaces enumeration: instead of comparing a movie to fifteen thousand others through brittle similarity rules, the system asks a single question — &lt;strong>how far apart are these two vectors?&lt;/strong>&lt;/p></description></item></channel></rss>