TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Variational Autoencoders are not autoencoders

66 pointsby ml_basicsover 6 years ago

3 comments

angel_jover 6 years ago
Pretty sure all this says is to minimize KL-Divergence instead of log-likelihood (for the encoder), or your latent variables are garbage. Judging by many examples I&#x27;ve seen of VAEs in ML, this is not news.<p>This post [0] does a good job showing the difference between latent variables for log-likelihood and KL losses.<p>[0]<a href="https:&#x2F;&#x2F;towardsdatascience.com&#x2F;intuitively-understanding-variational-autoencoders-1bfe67eb5daf" rel="nofollow">https:&#x2F;&#x2F;towardsdatascience.com&#x2F;intuitively-understanding-var...</a>
yorwbaover 6 years ago
Getting a resource exhaustion error from Namecheap, here&#x27;s a cache:<p><a href="http:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20190129144610&#x2F;http:&#x2F;&#x2F;paulrubenstein.co.uk&#x2F;variational-autoencoders-are-not-autoencoders&#x2F;" rel="nofollow">http:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20190129144610&#x2F;http:&#x2F;&#x2F;paulrubenst...</a>
no_identdover 6 years ago
What about variational homoencoders tho? See <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1807.08919" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1807.08919</a>