Using dimensionality reduction to exploit constraints in reinforcement learning.
Sebastian BitzerMatthew HowardSethu VijayakumarPublished in: IROS (2010)
Keyphrases
- dimensionality reduction
- reinforcement learning
- low dimensional
- data representation
- feature extraction
- high dimensional data
- principal component analysis
- manifold learning
- constraint satisfaction
- metric learning
- machine learning
- geometric constraints
- global constraints
- optimal policy
- reinforcement learning algorithms
- dimensionality reduction methods
- policy search
- principal components
- high dimensionality
- markov decision processes
- supervised learning
- state space
- feature space
- image processing
- genetic algorithm