Summary of Conformal Prediction Sets Can Cause Disparate Impact, by Jesse C. Cresswell et al.
Conformal Prediction Sets Can Cause Disparate Impact
by Jesse C. Cresswell, Bhargava Kumar, Yi Sui, Mouloud Belbahri
First submitted to arxiv on: 2 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 This paper introduces a statistically rigorous method for quantifying uncertainty in models called conformal prediction. The method produces sets of predictions with larger sets indicating more uncertainty. However, many applications require a single output rather than multiple predictions. To overcome this limitation, the authors suggest providing prediction sets to humans who then make informed decisions. They also investigate the fairness of outcomes across protected groups and propose using Equalized Set Sizes instead of Equalized Coverage. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making computers more unsure, which can be helpful in some situations. It’s like when you’re trying to decide what to wear and you want to consider multiple options. The authors developed a way for computers to do this too. They also looked at how this would affect different groups of people and found that it could make things worse if we don’t think about fairness. |
