The Importance of Sampling inMeta-Reinforcement Learning.
Bradly C. StadieGe YangRein HouthooftXi ChenYan DuanYuhuai WuPieter AbbeelIlya SutskeverPublished in: NeurIPS (2018)
Keyphrases
- reinforcement learning
- state space
- relative importance
- random sampling
- monte carlo
- function approximation
- learning algorithm
- sample size
- learning problems
- sampling strategies
- sampling strategy
- sampling algorithm
- reinforcement learning algorithms
- markov decision processes
- optimal policy
- neural network
- real time
- temporal difference
- evolutionary algorithm
- sampling methods
- action space
- machine learning
- temporal difference learning