<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>The semantic landscape paradigm for neural networks · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2307.09550/</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/2307.09550/feed.xml" rel="self" type="application/rss+xml" /><item><title>The semantic landscape paradigm for neural networks — mentioned by esafak</title><link>https://news.ycombinator.com/item?id=37078757</link><guid isPermaLink="false">urn:hackerpapers:mention:2307.09550:37078757</guid><pubDate>Thu, 10 Aug 2023 17:06:12 GMT</pubDate><description>&lt;p&gt;I haven&amp;#x27;t read the latest literature but my understanding is that &amp;quot;grokking&amp;quot; is the phase transition that occurs during the coalescing of islands of understanding (increasingly abstract features) that eventually form a pathway to generalization. And that this is something associated with over-parameterized models, which have the potential to learn multiple paths (explanations). https://en.wikipedia.org/wiki/Percolation_theory A relevant, recent paper I found from a quick search: The semantic landscape paradigm for neural networks (https://arxiv.org/abs/2307.09550)&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2307.09550/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item></channel></rss>
