Biased Gradient Estimate with Drastic Variance Reduction for Meta Reinforcement Learning.
Yunhao TangPublished in: ICML (2022)
Keyphrases
- variance reduction
- policy gradient
- gradient estimation
- reinforcement learning
- confidence intervals
- monte carlo
- sample size
- bias variance decomposition
- quasi monte carlo
- reinforcement learning algorithms
- function approximation
- importance sampling
- naive bayes classifier
- machine learning
- test set
- knn
- classification accuracy
- decision trees
- learning algorithm