A Variational Autoencoder is a type of likelihood-based generative model. It consists of an encoder, that takes in data $x$ as input and transforms this into a latent representation $z$, and a decoder, that takes a latent representation $z$ and returns a reconstruction $\hat{x}$. Inference is performed via variational inference to approximate the posterior of the model.
Source: Auto-Encoding Variational BayesPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Decoder | 55 | 8.33% |
Image Generation | 37 | 5.61% |
Disentanglement | 31 | 4.70% |
Denoising | 20 | 3.03% |
Quantization | 14 | 2.12% |
Language Modelling | 12 | 1.82% |
Text Generation | 12 | 1.82% |
Image Classification | 12 | 1.82% |
Time Series Analysis | 11 | 1.67% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |