Lazy-MDPs: Towards Interpretable Reinforcement Learning by Learning When to Act.
Alexis JacqJohan FerretOlivier PietquinMatthieu GeistPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- learning algorithm
- markov decision processes
- learning process
- learning systems
- online learning
- state space
- supervised learning
- learning problems
- dynamical systems
- reinforcement learning methods
- partially observable
- model free
- function approximation
- active learning
- learning tasks
- transfer learning
- reinforcement learning algorithms
- learning environment
- rl algorithms