Cross-Silo Federated Learning Across Divergent Domains with Iterative Parameter Alignment

arXiv:2311.04818v3 Announce Type: replace
Abstract: Learning from the collective knowledge of data dispersed across private sources can provide neural networks with enhanced generalization capabilities. Federated learning, a method for collaboratively training a machine learning model across remote clients, achieves this by combining client models via the orchestration of a central server. However, current approaches face two critical limitations: i) they struggle to converge when client domains are sufficiently different, and ii) current aggregation techniques produce an identical global model for …

abstract alignment arxiv capabilities client collective cs.cv cs.dc cs.lg data domains federated learning iterative knowledge machine machine learning machine learning model networks neural networks orchestration training type via