Optimizing Federated Learning Client Selection via Multi-Objective Contextual Bandits

TMLR Paper4812 Authors

Abstract: In the rapidly evolving field of Machine Learning (ML), Federated Learning (FL) emerges as an innovative approach for training models across distributed devices without centralizing raw data. However, FL faces significant challenges due to the heterogeneous nature of client devices, leading to non-IID data distributions and various resource constraints. Moreover, the inherent bandwidth limitations in decentralized settings necessitate the efficient use of both network and energy resources. Energy-efficient clients not only reduce the frequency of battery charging but also minimize data transmissions, thereby resulting in lower overall energy consumption during model training. This reduction in energy usage not only improves network efficiency but also contributes to environmental sustainability. To address these challenges, we introduce a novel solution, Pareto Contextual Zooming for Federated Learning (PCZFL), which treats the client selection problem in FL as a multi-objective bandit problem. Our method focuses on optimizing both global accuracy and energy efficiency in parallel. By dynamically adjusting client selection based on real-time accuracy and energy context, the proposed solution ensures effective participation while minimizing energy consumption. In addition, we provide theoretical analysis on both the regret bound and time complexity of our method. Extensive experiments demonstrate that PCZFL noticeably outperforms current state-of-the-art methods, offering a robust solution that balances the competing demands of accuracy and energy efficiency in FL deployments.

Submission Length: Regular submission (no more than 12 pages of main content)

Previous TMLR Submission Url: https://openreview.net/forum?id=0A8sG8LYbc

Changes Since Last Submission: Our previous submission was desk‐rejected for altering the template’s default fonts. We located and removed the \usepackage{times} directive from our source, recompiled with the original TMLR template fonts, and regenerated the PDF. The revised manuscript now fully conforms to TMLR’s formatting requirements.

Assigned Action Editor: ~Tian_Li1

Submission Number: 4812