Reinforcement Using Supervised Learning for Policy Generalization.
Julien LaumonierPublished in: AAAI (2007)
Keyphrases
- supervised learning
- reinforcement learning
- optimal policy
- unsupervised learning
- agent receives
- active learning
- policy search
- learning algorithm
- markov decision process
- learning machines
- learning problems
- action space
- action selection
- semi supervised learning
- supervised classification
- training data
- multiple instance learning
- control policies
- training examples
- markov decision processes
- inductive bias
- learning tasks
- supervised machine learning
- training set
- asymptotically optimal
- real time
- state dependent
- statistical learning
- training samples
- semi supervised
- machine learning
- function approximation
- decision problems
- temporal difference
- partially observable markov decision processes
- labeled data
- policy makers
- support vector machine
- dynamic programming
- decision trees
- policy gradient
- neural network
- database