Summary of Federated Fairness Analytics: Quantifying Fairness in Federated Learning, by Oscar Dilley et al.
Federated Fairness Analytics: Quantifying Fairness in Federated Learning
by Oscar Dilley, Juan Marcelo Parra-Ullauri, Rasheed Hussain, Dimitra Simeonidou
First submitted to arxiv on: 15 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Computer Science and Game Theory (cs.GT); Neural and Evolutionary Computing (cs.NE)
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 Federated Learning (FL) methodology, Federated Fairness Analytics, aims to address fairness challenges in FL by introducing a framework for measuring fairness. The approach defines four notions of fairness with novel metrics, leveraging techniques from Explainable AI (XAI), cooperative game-theory, and networking engineering. Experimental results demonstrate the impact of statistical heterogeneity and client participation on fairness-performance trade-offs, highlighting the potential benefits of fairness-conscious approaches like Ditto and q-FedAvg. The methodology enables FL practitioners to uncover insights into their system’s fairness at various levels of granularity, facilitating the development of fairer FL systems. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated Learning is a way for many devices or computers to learn together without sharing all their data. This helps keep personal information private. But it also raises questions about being fair. The paper proposes a new way to measure fairness in Federated Learning, called Federated Fairness Analytics. It defines four ways to think about fairness and shows how different methods can improve or worsen fairness-performance trade-offs. The results show that fairness is affected by things like data quality and which devices are involved. This research aims to help developers create fairer systems. |
Keywords
» Artificial intelligence » Federated learning
