ACE-RL-Checkers: Improving automatic case elicitation through knowledge obtained by reinforcement learning in player agents.
Henrique Castro NetoRita Maria da Silva JuliaPublished in: CIG (2015)
Keyphrases
- reinforcement learning
- multi agent
- learned knowledge
- learning agents
- complex domains
- action selection
- function approximation
- single agent
- multi agent systems
- optimal policy
- learning algorithm
- multi agent reinforcement learning
- learning agent
- multi agent environments
- multiple agents
- intelligent agents
- temporal difference
- reinforcement learning algorithms
- state space
- reinforcement learning agents
- learning process
- multiagent systems
- temporal difference learning
- reasoning process
- dynamic environments
- dynamic programming
- cooperative
- case base
- preference elicitation
- learning capabilities
- multiagent reinforcement learning
- direct policy search
- autonomous learning
- reinforcement learning methods
- partial knowledge
- model free
- markov decision processes
- autonomous agents
- knowledge base