Federated Active Learning Under Extreme Non-IID and Global Class Imbalance

2026-03-11

AI Summary

This work addresses the challenges of low labeling efficiency and degraded model performance in federated active learning under extreme non-IID settings with global class imbalance. The authors propose FairFAL, a novel framework that reveals—for the first time—the critical role of query models in enabling class-balanced sampling. FairFAL employs lightweight prediction discrepancy analysis to adaptively infer both global class imbalance and local-global distribution shifts, dynamically selecting between global or local query models accordingly. Integrating a prototype-guided pseudo-labeling mechanism with a two-stage uncertainty-diversity balanced sampling strategy based on k-center, FairFAL significantly improves model accuracy and class coverage on long-tailed heterogeneous data. Extensive experiments demonstrate consistent superiority over existing methods across five benchmark datasets.

Technology Category

• Machine Learning: Active Learning
• Search and Optimization: Sampling/Simulation-based Search
• Game Theory and Economic Paradigms: Fair Division

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
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Federated active learning (FAL) seeks to reduce annotation cost under privacy constraints, yet its effectiveness degrades in realistic settings with severe global class imbalance and highly heterogeneous clients. We conduct a systematic study of query-model selection in FAL and uncover a central insight: the model that achieves more class-balanced sampling, especially for minority classes, consistently leads to better final performance. Moreover, global-model querying is beneficial only when the global distribution is highly imbalanced and client data are relatively homogeneous; otherwise, the local model is preferable. Based on these findings, we propose FairFAL, an adaptive class-fair FAL framework. FairFAL (1) infers global imbalance and local-global divergence via lightweight prediction discrepancy, enabling adaptive selection between global and local query models; (2) performs prototype-guided pseudo-labeling using global features to promote class-aware querying; and (3) applies a two-stage uncertainty-diversity balanced sampling strategy with k-center refinement. Experiments on five benchmarks show that FairFAL consistently outperforms state-of-the-art approaches under challenging long-tailed and non-IID settings. The code is available at https://github.com/chenchenzong/FairFAL.

Problem

• Federated Active Learning
• Non-IID
• Class Imbalance
• Annotation Cost
• Heterogeneous Clients

Innovation

• Federated Active Learning
• Class Imbalance
• Non-IID
• Adaptive Querying
• Prototype-guided Pseudo-labeling