Learning the aerodynamic design of supercritical airfoils through deep reinforcement learning.
Runze LiYufei ZhangHaixin ChenPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- learning algorithm
- learning process
- learning problems
- learning systems
- supervised learning
- optimal policy
- learning agents
- hybrid learning
- learning design
- function approximation
- machine learning
- active learning
- multi agent systems
- bayesian networks
- eligibility traces
- policy search
- evolutionary learning
- previously learned
- case study
- learned knowledge
- controller design
- model free
- knowledge acquisition
- prior knowledge
- design principles
- state space
- online learning
- design process