首页>
外国专利>
DATA-BASED REINFORCEMENT LEARNING DEVICE FOR IMPROVING LIMIT RUN-OUT RATE AND METHOD THEREOF
DATA-BASED REINFORCEMENT LEARNING DEVICE FOR IMPROVING LIMIT RUN-OUT RATE AND METHOD THEREOF
展开▼
机译:基于数据的强化学习率极限提高率的学习装置及其方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
Disclosed is a data-based reinforcement learning device for increasing a limit run-out rate. According to the present invention, an agent (100) learns a reinforcement learning model so that a reward for an action selectable according to a current state in an arbitrary environment (200) is maximized, wherein a difference between a total fluctuation rate and an individual fluctuation rate that fluctuates depending on an individual action for each action is provided as the reward for the agent (100).;COPYRIGHT KIPO 2020
展开▼