<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Topos and Stacks of Deep Neural Networks · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2106.14587/</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/2106.14587/feed.xml" rel="self" type="application/rss+xml" /><item><title>Topos and Stacks of Deep Neural Networks — mentioned by ly3xqhl8g9</title><link>https://news.ycombinator.com/item?id=36278298</link><guid isPermaLink="false">urn:hackerpapers:mention:2106.14587:36278298</guid><pubDate>Sun, 11 Jun 2023 05:06:02 GMT</pubDate><description>&lt;p&gt;None of those mentioned are ecofascists, especially not Simondon [1], that was the point: the true radicals destroy the future worlds, not directly the present one. Alan Turing destroyed all the multiverse branches [2] in which we don&amp;#x27;t use computation. Norman Borlaug [3] destroyed all the multiverse branches in which we don&amp;#x27;t have high-yield wheat. Perhaps an even better parallel would have been Alexander Grothendieck [3], the mathematician of the 20th century (maybe even of the 21st century if concepts as the topos [4] are made into usable tools for deep neural networks [5]), but also a…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2106.14587/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>math.AT</category></item></channel></rss>
