Can Temporal-Difference and Q-Learning Learn Representation? A Mean-Field Theory.
Yufeng ZhangQi CaiZhuoran YangYongxin ChenZhaoran WangPublished in: CoRR (2020)
Keyphrases
- temporal difference
- function approximation
- td learning
- reinforcement learning
- function approximators
- temporal difference learning
- reinforcement learning algorithms
- action selection
- model free
- evaluation function
- temporal difference methods
- monte carlo
- policy iteration
- step size
- learning tasks
- td methods
- multi agent
- actor critic
- learning algorithm
- supervised learning
- multiscale
- policy evaluation
- neural network
- data mining
- decision making
- support vector machine
- fixed point
- radial basis function