Deep Belief Nets as Function Approximators for Reinforcement Learning.
Farnaz AbtahiIan R. FaselPublished in: Lifelong Learning (2011)
Keyphrases
- belief nets
- function approximators
- reinforcement learning
- function approximation
- temporal difference
- reinforcement learning algorithms
- logistic regression
- belief networks
- conditional probabilities
- neural network
- temporal difference methods
- state space
- learning tasks
- causal models
- continuous state and action spaces
- graphical structure
- markov decision processes
- model free
- optimal policy
- action space
- policy search
- decision trees
- learning algorithm
- graphical models
- markov decision problems
- transfer learning
- support vector
- machine learning