A New Policy Iteration Algorithm For Reinforcement Learning in Zero-Sum Markov Games.
Anna WinnickiR. SrikantPublished in: CoRR (2023)
Keyphrases
- markov games
- policy iteration algorithm
- markov decision processes
- reinforcement learning
- reinforcement learning algorithms
- policy iteration
- finite state
- markov decision process
- control problems
- state space
- optimal policy
- function approximation
- actor critic
- multiagent reinforcement learning
- stochastic games
- dynamic programming
- partially observable
- temporal difference
- temporal difference learning
- action space
- approximate dynamic programming
- learning algorithm
- reinforcement learning methods
- machine learning
- action selection
- infinite horizon
- optimal control
- multi agent
- average cost
- average reward
- function approximators
- learning automata
- model free
- incomplete information
- decision problems
- supervised learning