Federated Q-Learning: Linear Regret Speedup with Low Communication Cost.
Zhong ZhengFengyu GaoLingzhou XueJing YangPublished in: CoRR (2023)
Keyphrases
- communication cost
- distributed data
- sensor networks
- communication overhead
- data distribution
- low bandwidth
- reinforcement learning
- cooperative
- function approximation
- online learning
- processing cost
- reduce communication cost
- network size
- optimal policy
- lower bound
- learning rate
- model free
- wireless sensor networks
- data sets
- multi agent
- data availability
- low overhead
- learning algorithm