Evaluation of Batch-Mode Reinforcement Learning Methods for Solving DEC-MDPs with Changing Action Sets.
Thomas GabelMartin A. RiedmillerPublished in: EWRL (2008)
Keyphrases
- action sets
- dec mdps
- reinforcement learning
- batch mode
- mechanism design
- cooperative multiagent systems
- markov decision processes
- reinforcement learning algorithms
- finite state
- state space
- game theory
- incremental learning
- active learning
- multi agent
- optimal policy
- markov decision problems
- temporal difference
- dynamic programming
- multiagent systems
- lower bound
- machine learning
- learning algorithm
- multi agent systems
- control policy
- computationally expensive
- data mining
- multistage
- control system