Summary of What Hides Behind Unfairness? Exploring Dynamics Fairness in Reinforcement Learning, by Zhihong Deng et al.
What Hides behind Unfairness? Exploring Dynamics Fairness in Reinforcement Learning
by Zhihong Deng, Jing Jiang, Guodong Long, Chengqi Zhang
First submitted to arxiv on: 16 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Methodology (stat.ME)
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 The paper investigates the sources of unfairness in reinforcement learning (RL) problems involving sensitive attributes like race and gender. It proposes a novel notion called dynamics fairness, which captures inequality stemming from environmental dynamics. The authors introduce identification formulas to quantify this concept using data, and demonstrate its effectiveness through extensive experiments. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Reinforcement learning is used to solve sequential decision-making problems, but it’s important to make sure the decisions are fair. This means considering factors like race and gender. Previous studies have looked at different types of fairness, but they didn’t understand why unfairness happens in RL. The researchers in this study wanted to figure out where inequality comes from. They did this by looking at the data generation process and breaking down how sensitive attributes affect long-term well-being. They also came up with a new way to measure fairness called dynamics fairness. This takes into account changes in the environment, decisions made, and past events. The team tested their methods using experiments and found that they worked well. |
Keywords
» Artificial intelligence » Reinforcement learning » Stemming
