Q-Learning in Simulated Robotic Soccer - Large State Spaces and Incomplete Information.
Karl TuylsSam MaesBernard ManderickPublished in: ICMLA (2002)
Keyphrases
- incomplete information
- robotic soccer
- state space
- autonomous agents
- multiagent systems
- reinforcement learning algorithms
- reinforcement learning
- multi agent
- medium size
- dynamic programming
- partial information
- markov decision processes
- missing information
- optimal policy
- markov chain
- query answering
- function approximation
- multi agent systems
- complex domains
- dynamic environments
- continuous state spaces
- motion control
- open world
- repeated games
- knowledge base
- nash equilibria
- model free
- cooperative
- real robot
- null values
- planning problems
- learning algorithm