强化学习理论、算法及应用

强化学习（ｒｅｉｎｆｏｒｃｅｍｅｎｔ ｌｅａｒｎｉｎｇ）一词来自于行为心理学，这一理论把行为学习看成是反复试验的过程，从而把环境状态映射成相应的动作。首先全面地介绍了强化学习理论的主要算法，即瞬时差分法、Ｑ－学习算法及自适应启发评价算法；然后介绍了强化学习的应用情况；最后讨论了强化学习目前所要研究的问题。

Reinforcement Learning Theory,Algorithms and Its Application

The term,reinforcement learning,comes from behavior psychology that takes behavior leaming as trial and error,by which the states of environment are mapped into corresponding actions.First,the main algorithms,temporal difference, \%Q \%learning and adaptive heuristic critic,are roundly introduced.Then,the application of reinforcement leaming is presented.Finally,some present research projects of reinforcement learning are discussed.