AI RESEARCH
Client-Conditional Federated Learning via Local Training Data Statistics
arXiv CS.LG
•
ArXi:2603.11307v1 Announce Type: new Federated learning (FL) under data heterogeneity remains challenging: existing methods either ignore client differences (FedAvg), require costly cluster discovery (IFCA), or maintain per-client models (Ditto). All degrade when data is sparse or heterogeneity is multi-dimensional. We propose conditioning a single global model on locally-computed PCA statistics of each client's