Re-thinking Federated Active Learning based on Inter-class Diversity
This work addresses the challenge of efficiently utilizing unlabeled data in federated learning for decentralized clients, offering a robust solution to improve active learning performance in real-world scenarios.
The paper tackles the problem of selecting query samples in federated active learning by analyzing the impact of global and local inter-class diversity on two selector models, and proposes LoGo, a sampling strategy that integrates both models to handle varying heterogeneity and imbalance, achieving consistent outperformance over six baselines across 38 experimental settings.
Although federated learning has made awe-inspiring advances, most studies have assumed that the client's data are fully labeled. However, in a real-world scenario, every client may have a significant amount of unlabeled instances. Among the various approaches to utilizing unlabeled data, a federated active learning framework has emerged as a promising solution. In the decentralized setting, there are two types of available query selector models, namely 'global' and 'local-only' models, but little literature discusses their performance dominance and its causes. In this work, we first demonstrate that the superiority of two selector models depends on the global and local inter-class diversity. Furthermore, we observe that the global and local-only models are the keys to resolving the imbalance of each side. Based on our findings, we propose LoGo, a FAL sampling strategy robust to varying local heterogeneity levels and global imbalance ratio, that integrates both models by two steps of active selection scheme. LoGo consistently outperforms six active learning strategies in the total number of 38 experimental settings.