On the basis of problem formulation of robot action planning, a new method is proposed for it, through applying the explanation based learning to acquire searching strategies.
本文在对机器人动作规划进行问题描述的基础上，通过应用解释学习来获取搜索策略，为其提出了一种新的方法。
This paper discusses the learning requirements of an information society, the components of the method of learning by action and its advantages, and the significance of the method.
本文分析了信息社会对学习的要求，介绍了行动学习法的涵义及其优势，同时论述了行动学习法的重要意义。
Q-learning is a typical Reinforcement Learning (RL) method with a slow convergence speed especially as the scales of the state space and action space increase.
学习是一种典型的强化学习，其学习效率较低，尤其是当状态空间和决策空间较大时。
应用推荐