Communication-Efficient Federated AUC Maximization with Cyclic Client Participation

2026-01-04

Trans. Mach. Learn. Res.

AI Summary

This work addresses the challenges of communication efficiency and convergence in federated learning with periodic client participation for non-decomposable AUC optimization. The authors propose a communication-efficient federated algorithm for AUC maximization, handling both squared surrogate loss and general pairwise AUC loss. They establish, for the first time under this setting, theoretical guarantees on both communication and iteration complexity, and introduce the Polyak–Łojasiewicz (PL) condition to significantly accelerate convergence. Built upon a nonconvex–strongly-concave minimax optimization framework, the method integrates communication compression with pairwise loss optimization. Extensive experiments on image classification, medical imaging, and fraud detection tasks demonstrate its effectiveness, achieving a communication complexity of $\widetilde{O}(1/\varepsilon^{1/2})$, which improves upon existing approaches.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Non-convex Optimization
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Federated AUC maximization is a powerful approach for learning from imbalanced data in federated learning (FL). However, existing methods typically assume full client availability, which is rarely practical. In real-world FL systems, clients often participate in a cyclic manner: joining training according to a fixed, repeating schedule. This setting poses unique optimization challenges for the non-decomposable AUC objective. This paper addresses these challenges by developing and analyzing communication-efficient algorithms for federated AUC maximization under cyclic client participation. We investigate two key settings: First, we study AUC maximization with a squared surrogate loss, which reformulates the problem as a nonconvex-strongly-concave minimax optimization. By leveraging the Polyak-{\L}ojasiewicz (PL) condition, we establish a state-of-the-art communication complexity of $\widetilde{O}(1/\epsilon^{1/2})$ and iteration complexity of $\widetilde{O}(1/\epsilon)$. Second, we consider general pairwise AUC losses. We establish a communication complexity of $O(1/\epsilon^3)$ and an iteration complexity of $O(1/\epsilon^4)$. Further, under the PL condition, these bounds improve to communication complexity of $\widetilde{O}(1/\epsilon^{1/2})$ and iteration complexity of $\widetilde{O}(1/\epsilon)$. Extensive experiments on benchmark tasks in image classification, medical imaging, and fraud detection demonstrate the superior efficiency and effectiveness of our proposed methods.

Problem

• Federated AUC Maximization
• Cyclic Client Participation
• Communication Efficiency
• Non-decomposable Objective
• Imbalanced Data

Innovation

• Federated AUC Maximization
• Cyclic Client Participation
• Communication Efficiency
• Minimax Optimization
• Polyak-Łojasiewicz Condition