Learn how autoencoders compress data via bottleneck latent encodings
Covers VAE, VQ-VAE, VQ-VAE-2, and TD-VAE variants
See how latent embeddings enable search and data compression
Key quotes
“A nice byproduct is dimension reduction: the bottleneck layer captures a compressed latent encoding.”
Transcript
[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 representation can be used as en embedding vector in various applications (i.e. search), help data compression, or reveal the underlying data generative factors.
