Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning.
Noah Y. SiegelJost Tobias SpringenbergFelix BerkenkampAbbas AbdolmalekiMichael NeunertThomas LampeRoland HafnerNicolas HeessMartin A. RiedmillerPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- model free
- reinforcement learning algorithms
- function approximation
- state space
- partially observable
- prior knowledge
- optimal policy
- robotic control
- markov decision processes
- evolutionary algorithm
- machine learning
- real time
- active learning
- prior information
- decision making
- information retrieval
- learning capabilities
- function approximators
- policy search
- database