Leveraging feature communication in federated learning for remote sensing image classification
This work addresses efficiency and privacy challenges in federated learning for remote sensing applications, though it appears incremental with new communication strategies.
The study tackled improving federated learning for remote sensing image classification by introducing feature-centric communication, pseudo-weight amalgamation, and combined methods, resulting in accelerated convergence, enhanced privacy, and reduced network information exchange as demonstrated on two public datasets.
In the realm of Federated Learning (FL) applied to remote sensing image classification, this study introduces and assesses several innovative communication strategies. Our exploration includes feature-centric communication, pseudo-weight amalgamation, and a combined method utilizing both weights and features. Experiments conducted on two public scene classification datasets unveil the effectiveness of these strategies, showcasing accelerated convergence, heightened privacy, and reduced network information exchange. This research provides valuable insights into the implications of feature-centric communication in FL, offering potential applications tailored for remote sensing scenarios.