ATraDiff: Accelerating Online Reinforcement Learning with Imaginary Trajectories.
Qianlan YangYu-Xiong WangPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- online learning
- learning algorithm
- data sets
- genetic algorithm
- active learning
- state space
- batch mode
- reinforcement learning algorithms
- moving object trajectories
- database
- stationary camera
- online environment
- temporal difference learning
- online advertising
- function approximation
- temporal information
- optimal policy
- supervised learning
- multi agent
- real time