Decentralized Federated Policy Gradient with Byzantine Fault-Tolerance and Provably Fast Convergence.
Philip JordanFlorian GrötschlaFlint Xiaofeng FanRoger WattenhoferPublished in: AAMAS (2024)
Keyphrases
- fault tolerance
- fault tolerant
- distributed systems
- policy gradient
- load balancing
- function approximation
- actor critic
- peer to peer
- convergence rate
- gradient method
- reinforcement learning
- approximation methods
- optimal control
- multi agent
- digital libraries
- database replication
- failure recovery
- fault management
- mobile agents
- cooperative
- single point of failure
- worst case
- single agent
- computational complexity
- reinforcement learning algorithms
- neural network