Summary of Supervised Algorithmic Fairness in Distribution Shifts: a Survey, by Minglai Shao et al.
Supervised Algorithmic Fairness in Distribution Shifts: A Survey
by Minglai Shao, Dong Li, Chen Zhao, Xintao Wu, Yujie Lin, Qin Tian
First submitted to arxiv on: 2 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
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 surveys the field of fairness-aware machine learning under distribution shifts, addressing the challenge of maintaining equitable predictions when faced with changes in data distributions. Supervised models are trained on specific datasets but deployed in environments where data distributions may shift over time, leading to unfair predictions that disproportionately affect certain groups. The survey covers various types of distribution shifts and existing methods based on these shifts, including six commonly used approaches in the literature. Additionally, it lists publicly available datasets and evaluation metrics for empirical studies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Machine learning models can be biased when trained on one type of data but deployed in a different environment. This paper looks at how to keep machine learning fair even if the data changes. They explore what happens when data distributions change from training to deployment, which can lead to unfair predictions that affect certain groups more than others. The paper also talks about existing methods for fixing this problem and lists datasets and metrics used in research. |
Keywords
* Artificial intelligence * Machine learning * Supervised
