<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Restructuring Vector Quantization with the Rotation Trick · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2410.06424/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2410.06424/feed.xml" rel="self" type="application/rss+xml" /><item><title>Restructuring Vector Quantization with the Rotation Trick — mentioned by GistNoesis</title><link>https://news.ycombinator.com/item?id=47349270</link><guid isPermaLink="false">urn:hackerpapers:mention:2410.06424:47349270</guid><pubDate>Thu, 12 Mar 2026 11:40:36 GMT</pubDate><description>&lt;p&gt;I think it boils down to the alternate view of rotations as two successive reflections. You can then use householder matrix to avoid trigonometry. These geometric math tricks are sometimes useful for efficient computations. For example you can improve Vector-Quantization Variational AutoEncoder (VQ-VAE) using a rotation trick, and compute it efficiently without trigonometry using Householder matrix to find the optimal rotation which map one vector to the other. See section 4.2 of [1] The question why would someone avoid trigonometry instead of looking toward it is another one. Trigonometry…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2410.06424/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item><item><title>Restructuring Vector Quantization with the Rotation Trick — HN submission</title><link>https://news.ycombinator.com/item?id=45907718</link><guid isPermaLink="false">urn:hackerpapers:submission:2410.06424:45907718</guid><pubDate>Wed, 12 Nov 2025 22:25:25 GMT</pubDate><description>&lt;p&gt;Restructuring Vector Quantization with the Rotation Trick&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;1 point; 0 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2410.06424/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item></channel></rss>
