<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Tighter Variational Bounds are Not Necessarily Better · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1802.04537/</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/1802.04537/feed.xml" rel="self" type="application/rss+xml" /><item><title>Tighter Variational Bounds are Not Necessarily Better — mentioned by stochastic_monk</title><link>https://news.ycombinator.com/item?id=16625733</link><guid isPermaLink="false">urn:hackerpapers:mention:1802.04537:16625733</guid><pubDate>Tue, 20 Mar 2018 03:12:14 GMT</pubDate><description>&lt;p&gt;The equations on this page don&amp;#x27;t render for me on either Firefox or Chrome, making reading this a little jarring. See their paper [0] for what I think is easier to follow. *A more informative title would be something like &amp;quot;Improving Variational Autoencoders with improved Evidence-Based Lower Bound [ELBO] methods. The &amp;quot;What&amp;#x27;s wrong with VAEs?&amp;quot; title, followed by paragraphs explaining how VAEs work, made me wonder why I was reading it in the first place, but practically improving ELBO methods for VAE implementations is a meaningful contribution. [0] &amp;quot;Tighter Variational Bounds are Not…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1802.04537/"&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>
