Summary of Understanding Learning with Sliced-wasserstein Requires Rethinking Informative Slices, by Huy Tran et al.
Understanding Learning with Sliced-Wasserstein Requires Rethinking Informative Slices
by Huy Tran, Yikun Bai, Ashkan Shahbazi, John R. Hershey, Soheil Kolouri
First submitted to arxiv on: 16 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Applications (stat.AP); Computation (stat.CO); Machine Learning (stat.ML)
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 revisits the classic Sliced-Wasserstein distance (SWD) and introduces a rescaling approach to make each slice informative, simplifying it to a single global scaling factor. This modification allows SWD to match or surpass the performance of more complex variants in various machine learning tasks. The authors demonstrate the effectiveness of their method through extensive experiments across different learning tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper takes a classic distance metric, Sliced-Wasserstein, and makes it better by adjusting how it’s used. This helps make it work well for many common machine learning tasks. The team tested their idea on lots of different problems and showed that it can be just as good as more complicated methods. |
Keywords
* Artificial intelligence * Machine learning
