Decentralised Q-Learning for Multi-Agent Markov Decision Processes with a Satisfiability Criterion.
Keshav P. KevalVivek S. BorkarPublished in: CoRR (2023)
Keyphrases
- markov decision processes
- multi agent
- reinforcement learning
- markov games
- multi agent systems
- policy iteration
- state space
- optimal policy
- reinforcement learning algorithms
- finite state
- cooperative
- planning under uncertainty
- function approximation
- average reward
- dynamic programming
- partially observable
- model free
- optimality criterion
- single agent
- stochastic shortest path
- reward function
- finite horizon
- decision processes
- decision theoretic planning
- discounted reward
- stochastic games
- multiagent systems
- state and action spaces
- model based reinforcement learning
- infinite horizon
- learning algorithm
- computational complexity
- reachability analysis
- transition matrices
- action sets
- average cost
- machine learning
- discount factor
- state abstraction
- markov decision problems
- multiple agents
- multiagent reinforcement learning
- continuous state spaces
- policy evaluation
- optimal control
- action space