Eigensubspace of Temporal-Difference Dynamics and How It Improves Value Approximation in Reinforcement Learning.
Qiang HeTianyi ZhouMeng FangSetareh MaghsudiPublished in: ECML/PKDD (4) (2023)
Keyphrases
- temporal difference
- reinforcement learning
- policy evaluation
- function approximation
- td learning
- reinforcement learning algorithms
- model free
- evaluation function
- monte carlo
- temporal difference learning
- action selection
- function approximators
- approximation methods
- actor critic
- state space
- policy iteration
- step size
- temporal difference methods
- dynamical systems
- multi agent
- supervised learning
- approximate dynamic programming
- learning algorithm
- td methods
- machine learning
- queueing networks
- markov decision processes