An Architecture Built for Federated Learning: Addressing Data Heterogeneity through Adaptive Normalization-Free Feature Recalibration
TL;DR: We propose an architectural approach to solve performance degradation in non-IID FL. Leveraging Weight Standardization and Channel attention, we show our method to consistently outperform baselines in a wide array of scenarios, with minimal overhead.
Abstract: Federated learning is a decentralized collaborative training paradigm preserving stakeholders’ data ownership while improving performance and generalization. However, statistical heterogeneity among client datasets degrades system performance. To address this issue, we propose **Adaptive Normalization-free Feature Recalibration (ANFR)**, the first architecture-level approach to combat heterogeneous data in FL. ANFR leverages weight standardization to avoid mismatched client statistics and inconsistent averaging, ensuring robustness under heterogeneity, and channel attention to produce learnable scaling factors for feature maps, suppressing inconsistencies across clients due to heterogeneity. We demonstrate that this improves class selectivity and channel attention weight distribution, while working with any aggregation method, supporting both global and personalized FL, and adding minimal overhead. ANFR offers a novel and versatile approach to the challenge of statistical heterogeneity. Extensive experiments show ANFR consistently outperforms established baselines across various aggregation methods, datasets, and heterogeneity conditions.
Primary Area: Deep Learning->Everything Else
Keywords: Federated Learning
Submission Number: 9967
Loading