FedERL: Federated Efficient and Robust Learning for Common Corruptions

2025-08-24

AI Summary

In federated learning, clients often suffer from resource constraints and exhibit poor robustness against common data corruptions—such as noise, blur, and weather artifacts—while existing robust training methods impose prohibitive computational overhead, hindering practical deployment. To address this, we propose FedERL, the first framework achieving *zero client-side robustness overhead*: all robustness enhancement—including Data-Agnostic Robust Training (DART)—is performed exclusively on the server, without requiring clients to access raw data or execute any additional robust operations. DART introduces lightweight, data-agnostic adversarial augmentation and regularization on the server, balancing efficiency and corruption resilience. Experiments demonstrate that FedERL reduces average client inference time and energy consumption by over 60%, while attaining superior robust accuracy compared to state-of-the-art federated robust methods—particularly under severe resource constraints.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated learning (FL) accelerates the deployment of deep learning models on edge devices while preserving data privacy. However, FL systems face challenges due to client-side constraints on computational resources, and from a lack of robustness to common corruptions such as noise, blur, and weather effects. Existing robust training methods are computationally expensive and unsuitable for resource-constrained clients. We propose FedERL, federated efficient and robust learning, as the first work to explicitly address corruption robustness under time and energy constraints on the client side. At its core, FedERL employs a novel data-agnostic robust training (DART) method on the server to enhance robustness without access to the training data. In doing so, FedERL ensures zero robustness overhead for clients. Extensive experiments demonstrate FedERL's ability to handle common corruptions at a fraction of the time and energy cost of traditional robust training methods. In scenarios with limited time and energy budgets, FedERL surpasses the performance of traditional robust training, establishing it as a practical and scalable solution for real-world FL applications.

Problem

• Addresses client-side computational constraints in federated learning
• Enhances robustness against common corruptions like noise and blur
• Reduces time and energy costs for robust training methods

Innovation

• Server-side data-agnostic robust training
• Zero robustness overhead for clients
• Handles corruptions with reduced time energy