Reinforcement Learning for Cooperative Actions in a Partially Observable Multi-agent System.
Yuki TaniguchiTakeshi MoriShin IshiiPublished in: ICANN (1) (2007)
Keyphrases
- partially observable
- cooperative
- reinforcement learning
- multi agent systems
- multi agent
- state space
- markov decision processes
- partially observable domains
- partial observability
- partial observations
- decision problems
- dynamical systems
- reward function
- action models
- markov decision problems
- hidden state
- infinite horizon
- optimal policy
- game theory
- partially observable environments
- function approximation
- partially observable markov decision process
- belief space
- reinforcement learning algorithms
- initially unknown
- single agent
- dynamic programming
- model free
- orders of magnitude
- partially observable markov decision processes
- belief state
- fully observable
- action space
- temporal difference
- action selection
- optimal control