Low-Dimensional State and Action Representation Learning with MDP Homomorphism Metrics.
Nicolò BotteghiMannes PoelBeril SirmaçekChristoph BrunePublished in: CoRR (2021)
Keyphrases
- low dimensional
- reinforcement learning
- learning algorithm
- markov decision processes
- real time dynamic programming
- state space
- state action
- active learning
- learning process
- decision theoretic planning
- high dimensional
- supervised learning
- online learning
- manifold learning
- decision theoretic
- action sequences
- partial observations