The Difficulty of Passive Learning in Deep Reinforcement Learning.
Georg OstrovskiPablo Samuel CastroWill DabneyPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- learning algorithm
- learning process
- learning problems
- evolutionary learning
- function approximation
- knowledge acquisition
- machine learning
- state space
- learning mechanism
- deep learning
- unsupervised learning
- learning systems
- background knowledge
- action selection
- learning capabilities
- stochastic games
- state action
- autonomous learning
- prior knowledge
- eligibility traces