Fault-Tolerant Federated Reinforcement Learning with Theoretical Guarantee.
Flint Xiaofeng FanYining MaZhongxiang DaiWei JingCheston TanBryan Kian Hsiang LowPublished in: CoRR (2021)
Keyphrases
- fault tolerant
- theoretical guarantees
- reinforcement learning
- distributed systems
- fault tolerance
- policy iteration
- worst case
- function approximation
- markov decision processes
- load balancing
- optimal policy
- state machine
- state space
- model free
- temporal difference
- safety critical
- high availability
- learning algorithm
- interconnection networks
- machine learning
- digital libraries
- multi agent
- markov decision process
- dynamic programming
- optimal solution
- metadata