Summary of Lasers: Latent Space Encoding For Representations with Sparsity For Generative Modeling, by Xin Li et al.
LASERS: LAtent Space Encoding for Representations with Sparsity for Generative Modeling
by Xin Li, Anand Sarwate
First submitted to arxiv on: 16 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed paper relaxes the structural assumption of Vector Quantization (VQ) in variational autoencoders (VAEs) and generative adversarial networks (GANs). Instead, it assumes that the latent space can be approximated by a union of subspaces model, which is learned/updated during training. This approach leads to more expressive representations and better reconstruction quality at the cost of a small computational overhead for latent space computation. The results suggest that the true benefit of VQ might not come from discretizing the latent space, but rather from lossy compressing it. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper takes a different approach to learning compact latent space representations. Instead of using Vector Quantization (VQ), it uses a union of subspaces model to approximate the latent space. This allows for more expressive representations and better reconstruction quality. The results show that this approach is effective and can be used in place of VQ-based methods. |
Keywords
» Artificial intelligence » Latent space » Quantization
