中文摘要：

为了在智能学习和改变规则的情况下,在线最小二乘法支持向量机可以高效地估计值函数,采用了一种基于最小二乘支持向量机的新算法,通过汽车过山地实例证明了在线最小二乘法支持向量机的优越性,验证了该方法的可行性和有效性,利用最小二乘支持向量机通过一系列线性方程求解,使得在线应用成为可能.

英文摘要：

An algorithm called Least Squares Support Vector Machine （L,S-SVM） is proposed in this paper, LS-SVM is solved by solving a set of linear equations which makes online implementation feasible. The online LS-SVM can efficiently estimate the value functions whenever the agent learns and changes its policy. To illustrate the favorable performance of the online LS-SVM, it is applied to the Mountain-Car task, verify the feasibility of the presented method and effectiveness.