Summary of Achieving Group Fairness Through Independence in Predictive Process Monitoring, by Jari Peeperkorn and Simon De Vos
Achieving Group Fairness through Independence in Predictive Process Monitoring
by Jari Peeperkorn, Simon De Vos
First submitted to arxiv on: 6 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
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 A novel approach in machine learning aims to ensure group fairness in predictive process monitoring by predicting future outcomes without considering sensitive information. By using historical execution data, biases can be encoded into models that may perpetuate unfair behavior when deployed for new cases. To address this issue, researchers introduced metrics like and distribution-based alternatives to measure independence. A composite loss function combining binary cross-entropy and Wasserstein distribution-based loss is proposed to balance predictive performance and fairness while allowing customizable trade-offs. The effectiveness of these methods is demonstrated through a controlled experimental setup. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Predictive process monitoring tries to guess what will happen next in a process, like predicting the outcome of a case. Machine learning models are used for this task, but they can learn biases from historical data that might make them unfair. This research looks at how to make sure predictions don’t depend on sensitive information like race or gender. They use special metrics to measure fairness and propose a new way to train models that balances being fair with doing well. The results show that these methods work in controlled experiments. |
Keywords
» Artificial intelligence » Cross entropy » Loss function » Machine learning
