From Autoencoder to Beta-VAE
lilianweng.github.io- Category
- Other
- Type
- ARTICLE
- Builder
- @lilianweng
- Added
- Jul 21, 2026
About
[Updated on 2019-07-18: add a section on VQ-VAE & VQ-VAE-2 .] [Updated on 2019-07-26: add a section on TD-VAE .] Autocoder is invented to reconstruct high-dimensional data using a neural network model with a narrow bottleneck layer in the middle (oops, this is probably not true for Variational Autoencoder , and we will investigate it in details in later sections). A nice byproduct is dimension reduction: the bottleneck layer captures a compressed latent encoding. Such a low-dimensional represent
Why it made the leaderboard
A single, rigorous walkthrough of the autoencoder-to-VAE lineage — including VQ-VAE, VQ-VAE-2, and TD-VAE — that explains the math and design tradeoffs behind latent-variable generative models, useful for anyone building embeddings, compression, or generative components.
Comments (0)
No comments yet
Indexed by a proprietary survey. Corrections welcome.