The Nature of Temporal Difference Errors in Multi-step Distributional Reinforcement Learning.
Yunhao TangMark RowlandRémi MunosBernardo Ávila PiresWill DabneyMarc G. BellemarePublished in: CoRR (2022)
Keyphrases
- temporal difference
- multi step
- td learning
- reinforcement learning
- function approximation
- evaluation function
- model free
- reinforcement learning algorithms
- temporal difference learning
- policy evaluation
- policy iteration
- action selection
- monte carlo
- actor critic
- step size
- function approximators
- machine learning
- temporal difference methods
- state space
- knn
- markov decision processes
- k nearest neighbor
- supervised learning
- semi supervised
- reinforcement learning methods
- learning algorithm
- weight vector
- machine learning algorithms
- small number
- decision making