Continual Offline Reinforcement Learning via Diffusion-based Dual Generative Replay.
Jinmei LiuWenbin LiXiangyu YueShilin ZhangChunlin ChenZhi WangPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- generative model
- anisotropic diffusion
- state space
- model free
- unsupervised learning
- learning algorithm
- diffusion process
- markov decision processes
- reinforcement learning algorithms
- real time
- robotic control
- multi agent
- temporal difference
- control problems
- data driven
- supervised learning
- dynamic programming
- temporal difference learning
- reinforcement learning methods
- diffusion processes
- transfer learning
- denoising
- human brain
- active learning
- action selection
- image processing