Summary of Progressive Compression with Universally Quantized Diffusion Models, by Yibo Yang et al.
Progressive Compression with Universally Quantized Diffusion Models
by Yibo Yang, Justus C. Will, Stephan Mandt
First submitted to arxiv on: 14 Dec 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 This paper explores the potential of diffusion probabilistic models for progressive coding in generative modeling tasks. By optimizing a variational evidence lower bound (ELBO) on the data likelihood, these models can achieve state-of-the-art results in tasks such as image generation and inverse problem solving. The authors propose a new form of diffusion model with uniform noise in the forward process, which corresponds to the end-to-end compression cost using universal quantization. They demonstrate promising first results on image compression, achieving competitive rate-distortion and rate-realism results on a wide range of bit-rates with a single model. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at how a type of machine learning called diffusion models can be used to compress images in a way that gets better as more bits are sent. The authors make a new kind of diffusion model that uses random noise and show it works well for image compression, getting good results on different bit-rates with just one model. |
Keywords
» Artificial intelligence » Diffusion » Diffusion model » Image generation » Likelihood » Machine learning » Quantization
