Addressing Data Heterogeneity in Federated Learning with Adaptive Normalization-Free Feature Recalibration

2024-10-02

arXiv.org

AI Summary

To address the core challenge of degraded model performance caused by client statistical heterogeneity in federated learning, this paper proposes a lightweight, plug-and-play backbone enhancement method. The method introduces a novel normalization-agnostic feature recalibration mechanism that couples weight standardization with channel-wise attention—marking the first such design—thereby significantly improving class selectivity and robustness of attention distributions without batch normalization. It also achieves the first privacy-utility Pareto-optimal trade-off under differential privacy constraints (DP-SGD, ε = 2–4). Compatible with arbitrary aggregation strategies, the method supports both global and personalized federated learning paradigms. Extensive experiments demonstrate consistent and significant improvements over state-of-the-art methods across diverse datasets, heterogeneity settings, and aggregation algorithms, with negligible computational overhead and accuracy approaching that of non-private baseline models.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Federated learning is a decentralized collaborative training paradigm that preserves stakeholders' data ownership while improving performance and generalization. However, statistical heterogeneity among client datasets poses a fundamental challenge by degrading system performance. To address this issue, we propose Adaptive Normalization-free Feature Recalibration (ANFR), an architecture-level approach that combines weight standardization and channel attention. Weight standardization normalizes the weights of layers instead of activations. This is less susceptible to mismatched client statistics and inconsistent averaging, thereby more robust under heterogeneity. Channel attention produces learnable scaling factors for feature maps, suppressing those that are inconsistent between clients due to heterogeneity. We demonstrate that combining these techniques boosts model performance beyond their individual contributions, by enhancing class selectivity and optimizing channel attention weight distribution. ANFR operates independently of the aggregation method and is effective in both global and personalized federated learning settings, with minimal computational overhead. Furthermore, when training with differential privacy, ANFR achieves an appealing balance between privacy and utility, enabling strong privacy guarantees without sacrificing performance. By integrating weight standardization and channel attention in the backbone model, ANFR offers a novel and versatile approach to the challenge of statistical heterogeneity. We demonstrate through extensive experiments that ANFR consistently outperforms established baselines across various aggregation methods, datasets, and heterogeneity conditions.

Problem

• Addressing statistical heterogeneity in federated learning systems
• Combining weight standardization and channel attention techniques
• Improving model performance under data distribution inconsistencies

Innovation

• Weight standardization for client statistics robustness
• Channel attention for learnable feature scaling factors
• Integration enabling privacy-utility balance in FL