Reinforcement Learning-Based Client Selection and Trajectory Optimization for Multi-UAV Federated Learning
A proximal policy optimization (PPO)-based reinforcement learning model under a centralized training and distributed execution (CTDE) framework is proposed, which outperforms benchmarks by optimizing the global model performance and the energy consumption, simultaneously.
Wonseok Lee, Young Jeon, Ba Cao Nguyen et al.
· IEEE Transactions on Mobile... · 0 citations