5 ms·
Variational Autoencoders are not autoencoders
- yorwba 8y agoGetting a resource exhaustion error from Namecheap, here's a cache: http://web.archive.org/web/20190129144610/http://paulrubenstein.co.uk/variational-autoencoders-are-not-autoencoders/ http://web.archive.org/web/20190129144610/http://paulrubenst...
- angel_j 8y agoPretty 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've seen of VAEs in ML, this is not news. This post [0] does a good job showing the difference between latent variables for log-likelihood and KL losses. [0]https://towardsdatascience.com/intuitively-understanding-variational-autoencoders-1bfe67eb5daf https://towardsdatascience.com/intuitively-understanding-var...
- no_identd 8y agoWhat about variational homoencoders tho? See https://arxiv.org/abs/1807.08919 https://arxiv.org/abs/1807.08919