Hierarchical Federated Learning with Dynamic Clustering and Adaptive Regularization for Robust Infrastructure Inspection
For infrastructure inspection practitioners, this work addresses the critical problem of data heterogeneity in federated learning, enabling privacy-preserving collaboration across diverse structural types and imbalanced local datasets.
This paper tackles the challenge of 'double heterogeneity' in federated learning for structural health monitoring, where macro-level structural differences and micro-level data imbalances degrade model performance. The proposed hierarchical framework with dynamic clustering and adaptive regularization achieves robust and specialized diagnostic models, outperforming baselines on a large-scale real-world dataset.
The deployment of data-driven computer vision models for structural health monitoring (SHM) is heavily constrained by the data silo dilemma due to stringent privacy and security regulations. While federated learning (FL) offers a privacy-preserving collaborative alternative, its application to nationwide infrastructure networks is severely hindered by the challenge of ``double heterogeneity'': macro-level physical divergence across disparate structural types and micro-level statistical imbalances within local datasets. To overcome this challenge, this paper proposes a novel hierarchical federated learning framework. The framework orchestrates a synergistic two-tier optimization strategy. At the macro-level, a dynamic gradient-based clustering mechanism autonomously aggregates distributed clients into specialized expert groups based on their structural degradation trajectories, circumventing the need for prior geographical metadata. Concurrently, at the micro-level, an intra-cluster Dynamic Region-Adaptive Proximal Regularization (DRAPR) module computes a real-time statistical Non-IID Intensity Score for each client. By adaptively modulating a proximal penalty based on local label skewness and gradient divergence, DRAPR effectively calibrates local updates, mitigates client drift, and prevents the catastrophic forgetting of minority damage classes. Comprehensive evaluations on a large-scale, real-world structural inspection dataset demonstrate that the hierarchical integration of macro-clustering and micro-regularization successfully neutralizes dual-level heterogeneity, yielding highly robust and specialized diagnostic models for complex infrastructure inspection.