Dimensionality Reduction and Prioritized Exploration for Policy Search.
Marius MemmelPuze LiuDavide TateoJan PetersPublished in: CoRR (2022)
Keyphrases
- policy search
- dimensionality reduction
- reinforcement learning
- principal component analysis
- continuous state
- feature extraction
- high dimensional
- continuous action
- data points
- dynamic programming
- policy gradient
- feature selection
- principal components
- reinforcement learning algorithms
- feature space
- reward function
- utility function
- partially observable markov decision processes