Learning Dynamics and Generalization in Reinforcement Learning.
Clare LyleMark RowlandWill DabneyMarta KwiatkowskaYarin GalPublished in: CoRR (2022)
Keyphrases
- state space
- reinforcement learning
- dynamical systems
- partially observable
- state abstraction
- learning algorithm
- dynamic programming
- markov decision process
- optimal policy
- learning process
- supervised learning
- reinforcement learning methods
- learning problems
- learning capabilities
- learning agents
- machine learning
- learning systems
- learning machines
- recurrent networks
- temporal difference learning
- efficient learning
- mobile learning
- knowledge acquisition
- prior knowledge