Commit 7400b285 authored by PhilFischer's avatar PhilFischer
Browse files

Update README.md

parent 41527254
......@@ -43,13 +43,3 @@ The decoder models the conditional likelihood given the latent sample $`p(\mathb
![](img/architecture.png)
## TODO:
- [ ] Implement content loss using Euclidean distance or Wasserstein metric of an image representation network.
- [ ] Optimize latent space rank: 1D or 2D representation.
- [ ] Optimize latent space form: [Quantized Latent Space](https://arxiv.org/abs/1711.00937).
- [x] Implement style loss using an oracle classifier.
- [x] Prevent posterior collapse: β-VAE or [σ-VAE](https://arxiv.org/abs/2006.13202).
- [x] Optimize feature extraction head with transfer learning.
- [x] Optimize decoder distribution: Standard Gaussian, Scaled Gaussian, Beta, Kumaraswamy, **Discrete**.
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment