Replica Tree-based Federated Learning using Limited Data

no code implementations • 28 Dec 2023 • Ramona Ghilea, Islem Rekik

Furthermore, we leverage the hierarchical structure of the client network (both original and virtual), alongside the model diversity across replicas, and introduce a diversity-based tree aggregation, where replicas are combined in a tree-like manner and the aggregation weights are dynamically updated based on the model discrepancy.

Federated Learning Graph Generation +1