Summary of Regmixmatch: Optimizing Mixup Utilization in Semi-supervised Learning, by Haorong Han et al.
RegMixMatch: Optimizing Mixup Utilization in Semi-Supervised Learning
by Haorong Han, Jidong Yuan, Chixuan Wei, Zhongyang Yu
First submitted to arxiv on: 14 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); 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 paper explores the application of consistency regularization and pseudo-labeling in semi-supervised learning (SSL). While Mixup has been used effectively for consistency regularization, our findings suggest that it may actually degrade SSL performance by compromising the purity of artificial labels. To address this issue, we propose RegMixMatch, a novel framework that optimizes the use of Mixup with both high- and low-confidence samples in SSL. Our approach introduces semi-supervised RegMixup to address reduced artificial label purity and develops class-aware Mixup to reduce confirmation bias associated with low-confidence samples. Experimental results demonstrate state-of-the-art performance across various SSL benchmarks, including CORA, Citeseer, Pubmed, and Amazon-670k. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about making computers learn better without a lot of training data. They found that some ways of doing this might actually make it worse! To fix this, they came up with a new way to use a technique called Mixup to help computers learn from both labeled and unlabeled data. This new approach combines the best parts of different ideas to get even better results. The authors tested their method on several big datasets and found that it outperformed previous methods in many cases. |
Keywords
» Artificial intelligence » Regularization » Semi supervised
