Cumulative Utility Parity for Fair Federated Learning under Intermittent Client Participation

2026-02-14

AI Summary

This work addresses the challenge of fairness in federated learning under heterogeneous and intermittent client participation, where existing methods often overlook disparities in participation opportunities, leading to persistent underrepresentation of intermittently available clients. To mitigate this, the authors propose a cumulative utility fairness principle that introduces an availability-normalized cumulative utility metric, decoupling physical availability constraints from scheduling bias and instead focusing on the long-term utility each client accrues per participation opportunity. Building on this principle, they design a novel fairness-aware objective function and aggregation mechanism, explicitly accounting for non-IID temporal data distributions. Experimental results demonstrate that the proposed approach significantly improves long-term representational fairness on temporally skewed, non-IID federated benchmarks while maintaining model performance close to the optimal.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Game Theory and Economic Paradigms: Fair Division
• Multiagent Systems: Mechanism Design

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In real-world federated learning (FL) systems, client participation is intermittent, heterogeneous, and often correlated with data characteristics or resource constraints. Existing fairness approaches in FL primarily focus on equalizing loss or accuracy conditional on participation, implicitly assuming that clients have comparable opportunities to contribute over time. However, when participation itself is uneven, these objectives can lead to systematic under-representation of intermittently available clients, even if per-round performance appears fair. We propose cumulative utility parity, a fairness principle that evaluates whether clients receive comparable long-term benefit per participation opportunity, rather than per training round. To operationalize this notion, we introduce availability-normalized cumulative utility, which disentangles unavoidable physical constraints from avoidable algorithmic bias arising from scheduling and aggregation. Experiments on temporally skewed, non-IID federated benchmarks demonstrate that our approach substantially improves long-term representation parity, while maintaining near-perfect performance.

Problem

• federated learning
• fairness
• intermittent participation
• client heterogeneity
• cumulative utility

Innovation

• cumulative utility parity
• fair federated learning
• intermittent client participation
• availability-normalized utility
• non-IID federated learning