Summary of Bandit Learning in Matching Markets: Utilitarian and Rawlsian Perspectives, by Hadi Hosseini and Duohan Zhang
Bandit Learning in Matching Markets: Utilitarian and Rawlsian Perspectives
by Hadi Hosseini, Duohan Zhang
First submitted to arxiv on: 30 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Science and Game Theory (cs.GT)
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 research paper proposes novel algorithms for two-sided matching markets, which have significant impacts in various real-world applications such as school choice, medical residency placement, and recommender systems. The traditional models assume known preferences, but in modern markets, preferences are unknown and must be learned. The authors adopt a welfarist approach to optimize matching for both sides of the market using two metrics: Utilitarian welfare and Rawlsian welfare, while maintaining market stability. They propose algorithms based on epoch Explore-Then-Commit (ETC) and analyze their regret bounds. Simulated experiments are conducted to evaluate both welfare and market stability. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about matching markets, which help people find good matches in things like schools or jobs. Right now, most models assume we know what people want ahead of time, but that’s not always true. In this case, a company might not know what kind of employees they need before interviewing candidates. Researchers have tried to solve this problem by treating matching markets like a “multi-armed bandit” game, which helps one side of the market get good matches. But these solutions often don’t work as well for the other side. This paper tries to fix that by making sure both sides of the market are happy and stable. |
