Entropy regularized actor-critic based multi-agent deep reinforcement learning for stochastic games.
Dong HaoDongcheng ZhangQi ShiKai LiPublished in: Inf. Sci. (2022)
Keyphrases
- stochastic games
- actor critic
- reinforcement learning
- multi agent
- reinforcement learning algorithms
- rl algorithms
- average reward
- single agent
- policy gradient
- markov decision processes
- multiagent reinforcement learning
- temporal difference
- model free
- state space
- function approximation
- state action
- cooperative
- learning automata
- learning agent
- policy iteration
- optimal policy
- multi agent systems
- learning algorithm
- approximate dynamic programming
- reinforcement learning methods
- temporal difference learning
- optimal control
- neuro fuzzy
- nash equilibria
- reward function
- least squares
- learning capabilities
- objective function
- autonomous agents
- multiple agents
- action selection
- gradient method
- long run
- imperfect information
- action space
- markov decision process
- supervised learning
- worst case