In order to solve both of the "curse of dimensionality" and slow convergence speed problem,a reward optimization method based on action sub-rewards in hierarchical reinforcement learning was proposed.
针对强化学习的“维数灾”问题和算法收敛速度过慢的困难,提出了一种基于动作分值的分层强化学习奖赏优化方法。
参考来源 - 强化学习维数灾问题解决方法研究·2,447,543篇论文数据，部分数据来源于NoteExpress
游戏玩家们必须尽可能的开快一点，更时髦抢眼一点，盖时髦的动作能获得荣誉分值，在职业赛中可是很有用的哦。
Gamers have to drive fast as well as drive in style, as stylish moves will earn them Kudos points, which will be helpful in the career.
中国女单难度动作的分值选择0.5和0.6分值的占多数，缺乏高分值难度动作，混双项目与其它国家差距不明显；
Chinese individual women chooses most of 0.5and 0.6 to difficulty value, and is lack of high value difficulty, mixed pair has small gap to other countries;
应用推荐