Reinforcement Learning with Factored States and Actions.
Brian SallansGeoffrey E. HintonPublished in: J. Mach. Learn. Res. (2004)
Keyphrases
- perceptual aliasing
- reinforcement learning
- state action space
- function approximation
- state space
- state action
- initial state
- action selection
- factored markov decision processes
- markov decision problems
- action space
- partially observable
- optimal policy
- state transitions
- state transition
- state and action spaces
- goal state
- state variables
- model free
- state information
- state abstraction
- reinforcement learning algorithms
- action sequences
- heuristic search
- learning algorithm
- multi agent
- actor critic
- reward function
- evaluation function
- optimal control
- markov decision processes
- fully observable
- temporal difference
- reinforcement learning methods
- partial observability
- belief state
- markov chain
- machine learning
- action models
- search space
- partially observable domains
- reward signal
- transfer learning
- learning agent