Model-Based Multi-Agent RL in Zero-Sum Markov Games with Near-Optimal Sample Complexity.
Kaiqing ZhangSham M. KakadeTamer BasarLin F. YangPublished in: CoRR (2020)
Keyphrases
- markov games
- sample complexity
- multiagent reinforcement learning
- multi agent
- reinforcement learning algorithms
- reinforcement learning
- model free
- markov decision processes
- learning problems
- learning algorithm
- markov decision process
- stochastic games
- theoretical analysis
- multiagent systems
- supervised learning
- special case
- active learning
- control problems
- cooperative
- upper bound
- generalization error
- state space
- lower bound
- temporal difference
- function approximation
- optimal policy
- sample size
- multi agent systems
- machine learning
- finite state
- nash equilibrium
- autonomous agents
- data sets
- partially observable
- training examples
- dynamic programming
- model selection