Summary of Fairness-aware Multiobjective Evolutionary Learning, by Qingquan Zhang and Jialin Liu and Xin Yao
Fairness-aware Multiobjective Evolutionary Learning
by Qingquan Zhang, Jialin Liu, Xin Yao
First submitted to arxiv on: 27 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 proposed MOEL framework trains machine learning models that consider multiple conflicting objectives, including accuracy and fairness measures. While previous approaches construct a static representative set of fairness measures before model training, this paper suggests adapting the measure set online during training. The framework achieves state-of-the-art performance on 12 benchmark datasets for mitigating unfairness in terms of both accuracy and fairness metrics, using only dynamically selected optimisation objectives. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research trains machine learning models that are fairer by considering multiple goals, like being accurate and treating everyone equally. Instead of choosing a set of fairness measures before training the model, this paper suggests changing the measures during training to make them better. The results show that this approach works really well on many different datasets. |
Keywords
* Artificial intelligence * Machine learning
