Understanding Deep Neural Function Approximation in Reinforcement Learning via $\epsilon$-Greedy Exploration.
Fanghui LiuLuca VianoVolkan CevherPublished in: NeurIPS (2022)
Keyphrases
- function approximation
- reinforcement learning
- temporal difference
- function approximators
- exploration exploitation tradeoff
- model free
- temporal difference learning
- mountain car
- temporal difference learning algorithms
- dynamic programming
- state action space
- reinforcement learning algorithms
- action selection
- neural network
- learning tasks
- radial basis function
- optimal control
- machine learning
- tile coding
- learning algorithm
- data mining
- search space
- learning process
- state space
- temporal difference methods
- td learning
- transfer learning
- reinforcement learning problems
- supervised learning
- policy gradient
- image classification
- optimal policy
- feature selection