Optimal Policy of Multiplayer Poker via Actor-Critic Reinforcement Learning.
Daming ShiXudong GuoYi LiuWenhui FanPublished in: Entropy (2022)
Keyphrases
- optimal policy
- actor critic
- reinforcement learning
- policy iteration
- imperfect information
- average reward
- markov decision processes
- temporal difference
- state space
- decision problems
- reinforcement learning algorithms
- approximate dynamic programming
- finite horizon
- infinite horizon
- policy gradient
- function approximation
- multistage
- monte carlo
- learning algorithm
- finite state
- long run
- machine learning
- model free
- dynamic programming
- game playing
- stochastic games
- markov decision process
- initial state
- rl algorithms
- partially observable markov decision processes
- average cost
- sufficient conditions
- supervised learning
- partially observable
- reinforcement learning methods
- computer games
- learning problems
- multi agent
- optimal control
- markov decision problems
- function approximators
- state action
- nash equilibrium
- decision making