Communication-Efficient Federated Learning Based on Explanation-Guided Pruning for Remote Sensing Image Classification
This work addresses communication efficiency for operational applications in remote sensing with restricted bandwidth, though it is incremental as it builds on existing pruning methods with a novel guidance approach.
The paper tackles the high communication overhead in federated learning for remote sensing image classification by introducing an explanation-guided pruning strategy, which reduces the number of shared model updates by 30% while improving the global model's generalization ability on the BigEarthNet-S2 dataset.
Federated learning (FL) is a decentralized machine learning paradigm in which multiple clients collaboratively train a global model by exchanging only model updates with the central server without sharing the local data of the clients. Due to the large volume of model updates required to be transmitted between clients and the central server, most FL systems are associated with high transfer costs (i.e., communication overhead). This issue is more critical for operational applications in remote sensing (RS), especially when large-scale RS data is processed and analyzed through FL systems with restricted communication bandwidth. To address this issue, we introduce an explanation-guided pruning strategy for communication-efficient FL in the context of RS image classification. Our pruning strategy is defined based on the layer-wise relevance propagation (LRP) driven explanations to: 1) efficiently and effectively identify the most relevant and informative model parameters (to be exchanged between clients and the central server); and 2) eliminate the non-informative ones to minimize the volume of model updates. The experimental results on the BigEarthNet-S2 dataset demonstrate that our strategy effectively reduces the number of shared model updates, while increasing the generalization ability of the global model. The code of this work is publicly available at https://git.tu-berlin.de/rsim/FL-LRP.