Guiding Online Reinforcement Learning with Action-Free Offline Pretraining.
Deyao ZhuYuhui WangJürgen SchmidhuberMohamed ElhoseinyPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- real time
- action selection
- online learning
- partially observable domains
- state action
- optimal policy
- website
- policy search
- machine learning
- function approximation
- optimal control
- robotic control
- online environment
- batch mode
- action space
- reinforcement learning algorithms
- fitted q iteration
- human actions
- dynamic environments
- state space
- dynamic programming
- learning process
- multi agent
- computer vision
- genetic algorithm
- data sets