Nearly Optimal Regret for Learning Adversarial MDPs with Linear Function Approximation.
Jiafan HeDongruo ZhouQuanquan GuPublished in: CoRR (2021)
Keyphrases
- function approximation
- reinforcement learning
- learning tasks
- function approximators
- learning algorithm
- temporal difference methods
- learning process
- supervised learning
- markov decision processes
- temporal difference learning algorithms
- temporal difference learning
- pattern recognition
- policy evaluation
- td learning
- multi agent
- machine learning
- optimal policy
- radial basis function
- dynamic programming
- optimal control
- model free
- learning agent
- reinforcement learning problems