Policy Diagnosis via Measuring Role Diversity in Cooperative Multi-agent RL.
Siyi HuChuanlong XieXiaodan LiangXiaojun ChangPublished in: ICML (2022)
Keyphrases
- cooperative multi agent
- optimal policy
- reinforcement learning
- action selection
- markov decision processes
- markov decision process
- medical diagnosis
- policy search
- fault diagnosis
- state space
- markov decision problems
- model based diagnosis
- control policy
- action space
- multi agent
- control policies
- policy iteration
- learning algorithm
- optimal control
- function approximation
- reinforcement learning algorithms
- reward function
- model free
- infinite horizon
- fault detection
- policy gradient
- dynamic programming
- neural network