Summary of Selective Classification Under Distribution Shifts, by Hengyue Liang et al.
Selective Classification Under Distribution Shifts
by Hengyue Liang, Le Peng, Ju Sun
First submitted to arxiv on: 8 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); 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 A novel framework for generalized selective classification (GSC) is proposed to address distribution shifts in high-stakes scenarios. The GSC approach extends existing SC methods by considering label-shifted, covariate-shifted, and in-distribution samples. Two novel margin-based score functions are introduced for non-training-based confidence estimation on deep learning classifiers. Experimental results demonstrate the effectiveness and reliability of these scores compared to existing methods across various classification tasks and DL models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to help computers make better predictions is being developed. Sometimes, computers can’t be sure what they’re supposed to predict, so this new method helps them decide not to make a prediction at all if it’s likely to be wrong. This is important for situations where the computer’s mistakes could have big consequences. The team behind this project created two new ways for computers to measure how confident they are in their predictions, and these methods work better than what was available before. |
Keywords
» Artificial intelligence » Classification » Deep learning
