<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Change Actions: Models of Generalised Differentiation · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1902.05465/</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/1902.05465/feed.xml" rel="self" type="application/rss+xml" /><item><title>Change Actions: Models of Generalised Differentiation — mentioned by bmc7505</title><link>https://news.ycombinator.com/item?id=21439420</link><guid isPermaLink="false">urn:hackerpapers:mention:1902.05465:21439420</guid><pubDate>Mon, 04 Nov 2019 06:27:27 GMT</pubDate><description>&lt;p&gt;Thanks for checking out our project! Completely understand where you&amp;#x27;re coming from, I also struggled with imagining how differentiable programs could work, after all, computer programs implement discrete functions! When building it, I learned two surprising facts: (1) We can perform gradient descent on integer functions! [1] There are rigorous assumptions for differentiability (e.g. Lipschitz-continuity, β-smoothness), but most often, these don&amp;#x27;t apply in the real world. Although many discrete optimization problems can relaxed to semi-differentiable ones, even if a function is not strictly…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1902.05465/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LO</category></item></channel></rss>
