Reinforcement Learning of Perceptual Classes using Q Learning Updates.
Sébastien JodogneJustus H. PiaterPublished in: Artificial Intelligence and Applications (2005)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- state space
- model free
- learning algorithm
- optimal policy
- multi agent
- stochastic approximation
- markov decision processes
- action selection
- continuous state and action spaces
- reinforcement learning methods
- machine learning
- multi agent reinforcement learning
- control problems
- temporal difference learning
- state action space
- human visual system
- state abstraction
- low level
- dynamic programming
- function approximators
- learning agent
- temporal difference
- learning classifier systems
- human perception
- learning agents
- supervised learning
- class labels
- monte carlo
- continuous state spaces
- policy search
- relational reinforcement learning
- agent receives
- cooperative