Summary of Fair Classification with Partial Feedback: An Exploration-based Data Collection Approach, by Vijay Keswani et al.
Fair Classification with Partial Feedback: An Exploration-Based Data Collection Approach
by Vijay Keswani, Anay Mehrotra, L. Elisa Celis
First submitted to arxiv on: 17 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); 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 presents an innovative approach to training classifiers using available data, addressing the issue of biased training datasets. The method utilizes exploration strategies to collect outcome data about subpopulations that would otherwise be ignored, ensuring that all sub-populations are explored, and the fraction of false positives is bounded. This approach comes with guarantees that the trained classifier converges to a desired classifier, allowing for improved learning guarantees and context-specific group fairness properties. The authors evaluate their method on real-world datasets, showing consistent improvements in the quality of collected outcome data and the fraction of true positives for all groups, with only a small reduction in predictive utility. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us make better predictions by using available data to train classifiers. Normally, we can’t see what happens to people who are mistakenly left out of our system. This approach allows us to learn from those people and make sure our classifier is fair for everyone. The method uses strategies to collect more information about groups that were previously ignored. This helps ensure that all groups are treated equally and that the classifier makes accurate predictions. |
