Towards Learning Abstract Representations for Locomotion Planning in High-dimensional State Spaces.
Tobias KlamtSven BehnkePublished in: CoRR (2019)
Keyphrases
- high dimensional
- learning systems
- state space
- learning algorithm
- learning process
- prior knowledge
- active learning
- domain independent
- reinforcement learning
- action selection
- heuristic search
- online learning
- knowledge acquisition
- dimensionality reduction
- nearest neighbor
- degrees of freedom
- planning problems
- decision theoretic
- learning environment
- stochastic domains