Abstraction for Deep Reinforcement Learning.
Murray ShanahanMelanie MitchellPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- state abstraction
- function approximation
- high level
- markov decision processes
- learning algorithm
- temporal abstractions
- state space
- machine learning
- decision theoretic planning
- optimal policy
- robotic control
- multi agent
- reinforcement learning algorithms
- optimal control
- model free
- control problems
- function approximators
- data abstraction
- stochastic approximation
- data sets
- artificial neural networks
- learning process
- transfer learning
- learning capabilities
- monte carlo
- temporal difference learning
- decision making
- relational reinforcement learning
- dynamic programming