Theoretically Principled Deep RL Acceleration via Nearest Neighbor Function Approximation.
Junhong ShenLin F. YangPublished in: CoRR (2021)
Keyphrases
- function approximation
- nearest neighbor
- reinforcement learning
- tile coding
- k nearest neighbor
- model free
- temporal difference learning
- temporal difference
- knn
- temporal difference learning algorithms
- radial basis function
- training set
- high dimensional
- data points
- learning tasks
- temporal difference methods
- reinforcement learning algorithms
- function approximators
- rl algorithms
- exploration exploitation tradeoff
- state space
- learning problems
- optimal policy
- policy gradient
- td methods
- semi supervised
- decision trees
- machine learning