<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>When Does Preconditioning Help or Hurt Generalization? · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2006.10732/</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/2006.10732/feed.xml" rel="self" type="application/rss+xml" /><item><title>When Does Preconditioning Help or Hurt Generalization? — mentioned by vladf</title><link>https://news.ycombinator.com/item?id=47256769</link><guid isPermaLink="false">urn:hackerpapers:mention:2006.10732:47256769</guid><pubDate>Thu, 05 Mar 2026 02:33:19 GMT</pubDate><description>&lt;p&gt;That still looks like a “converge faster” paper. https://arxiv.org/abs/2006.10732 The above provides a nuanced theoretical view. GD inductive bias is probably better unless your model is misspecified&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2006.10732/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>stat.ML</category></item></channel></rss>
