Federated Unlearning: a Perspective of Stability and Fairness

Author:

Jiaqi Shao, Tao Lin, Xuanyu Cao, Bing Luo

Keyword:

Computer Science, Artificial Intelligence, Artificial Intelligence (cs.AI)

journal:

date:

2024-02-02 00:00:00

Abstract

This paper explores the multifaceted consequences of federated unlearning (FU) with data heterogeneity. We introduce key metrics for FU assessment, concentrating on verification, global stability, and local fairness, and investigate the inherent trade-offs. Furthermore, we formulate the unlearning process with data heterogeneity through an optimization framework. Our key contribution lies in a comprehensive theoretical analysis of the trade-offs in FU and provides insights into data heterogeneity's impacts on FU. Leveraging these insights, we propose FU mechanisms to manage the trade-offs, guiding further development for FU mechanisms. We empirically validate that our FU mechanisms effectively balance trade-offs, confirming insights derived from our theoretical analysis.

PDF: Federated Unlearning: a Perspective of Stability and Fairness.pdf