Summary of Puffle: Balancing Privacy, Utility, and Fairness in Federated Learning, by Luca Corbucci et al.
PUFFLE: Balancing Privacy, Utility, and Fairness in Federated Learning
by Luca Corbucci, Mikko A Heikkila, David Solans Noguero, Anna Monreale, Nicolas Kourtellis
First submitted to arxiv on: 21 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); 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 tackles the challenging problem of developing Machine Learning (ML) models that balance fairness, privacy, and utility. The authors highlight the importance of considering all three factors simultaneously, as many current efforts focus on just two of these aspects. In Federated Learning (FL), decentralization and variations in data distributions further complicate this ethical trade-off. To address this issue, the researchers introduce PUFFLE, a high-level parameterized approach that explores the balance between utility, privacy, and fairness in FL scenarios. Experimental results show that PUFFLE can effectively reduce model unfairness up to 75% while maintaining strict privacy guarantees, with a maximum reduction in utility of 17%. The authors demonstrate PUFFLE’s effectiveness across diverse datasets, models, and data distributions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making sure machine learning models are fair, private, and useful. It’s like trying to find the perfect balance between three important things. Right now, many efforts only focus on two of these aspects, which isn’t good enough. The authors introduce a new approach called PUFFLE that helps us figure out how to balance all three. They tested it with different datasets and models and found that it can make models up to 75% fairer while still being useful and keeping people’s personal information private. |
Keywords
» Artificial intelligence » Federated learning » Machine learning
