Open access
2026
A Reinforcement Learning–Driven Latency Optimization Framework for Heterogeneous Federated Learning on Edge Devices
A server-centric orchestration framework that jointly models and optimizes the entire FL pipeline through a three-stage latency model, multi-factor client selection, hierarchical aggregation, and reinforcement-learning-driven decision making is proposed.
De-Sheng Sun, Li-Xing Chen, Qi-Lin Wei et al.
· IEEE Access · 0 citations