Summary of Variance Reduction Of Diffusion Model’s Gradients with Taylor Approximation-based Control Variate, by Paul Jeha et al.
Variance reduction of diffusion model’s gradients with Taylor approximation-based control variate
by Paul Jeha, Will Grathwohl, Michael Riis Andersen, Carl Henrik Ek, Jes Frellsen
First submitted to arxiv on: 22 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 abstract presents a novel approach to reduce the high variance of score-based models, which are trained with denoising score matching for generating high-dimensional data. By introducing a control variate derived from a k-th order Taylor expansion of the training objective and its gradient, the authors aim to improve optimisation. The method is demonstrated empirically on low-dimensional problems and its effectiveness is studied on larger problem settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research improves score-based models for generating high-dimensional data by reducing their high variance. These models are trained with denoising score matching, which makes them very effective. However, this training process can be tricky because of the high variance. The authors found a way to fix this issue by using something called a control variate. This helps the model learn better and faster. |
