Summary of Falcon: Fair Active Learning Using Multi-armed Bandits, by Ki Hyun Tae et al.
Falcon: Fair Active Learning using Multi-armed Bandits
by Ki Hyun Tae, Hantian Zhang, Jaeyoung Park, Kexin Rong, Steven Euijong Whang
First submitted to arxiv on: 23 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 proposes Falcon, a scalable fair active learning framework that improves machine learning model fairness via strategic sample selection. The framework adopts a data-centric approach to identify samples from “target groups” (e.g., female-positive) that are most informative for improving fairness, given a user-specified group fairness measure. To handle the challenge of selecting samples without ground truth labels, the authors propose a novel trial-and-error method and observe the trade-off between informativeness and postpone rate, which varies per dataset. The optimal balance is captured as policies and selected using adversarial multi-armed bandit methods for computational efficiency and theoretical guarantees. Experimental results show that Falcon significantly outperforms existing fair active learning approaches in terms of fairness and accuracy, with a maximum fairness score 1.8-4.5x higher than the second-best results. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making machine learning models fair by choosing which data to use when training them. The problem is that if we only use certain types of data, our model might not be fair to other groups. To fix this, the authors created a new way to choose data that takes into account fairness and tries to balance accuracy with fairness. |
Keywords
* Artificial intelligence * Active learning * Machine learning
