巴韦利埃表示:“令人感到吃惊的是,靠玩动作游戏提高的概率推理能力并不仅限于游戏,同时也可用于完成与游戏无关的更为鼓噪的任务。”
“What’s surprising in our study is that action games improved probabilistic inference not just for the act of gaming, but for unrelated and rather dull tasks,” Bavelier says.
论文提出一种模糊强化学习算法,通过模糊推理系统将连续的状态空间映射到连续的动作空间,然后通过学习得到一个完整的规则库。
In this paper, we propose a fuzzy reinforcement algorithm, which map continuous state Spaces to continuous action Spaces by fuzzy inference system and then learn a rule base.
首先,提出一种模糊Q学习算法,通过模糊推理系统将连续的状态空间映射到连续的动作空间,然后通过学习得到一个完整的规则库。
A fuzzy Q learning algorithm is proposed in this dissertation, which map continuous state Spaces to continuous action Spaces by fuzzy inference system and then learn a rule base.
规则中受推理机测试的条件部分,若前提为真,则采取规则结论所规定的相应动作。
The condition portion of a rule that is tested by the inference engine. If the premise is found to be true, the corresponding action specified in the rule conclusion is taken.
规则中受推理机测试的条件部分,若前提为真,则采取规则结论所规定的相应动作。
The condition portion of a rule that is tested by the inference engine. If the premise is found to be true, the corresponding action specified in the rule conclusion is taken.
应用推荐