基于支持向量机的离散线性微分方程求解方法

Support Vector Machines Based Method to Solve Discrete Linear Differential Equations

支持向量机(Support Vector Machines, SVMs)在逼近问题的求解上展现出了良好的有效性和可行性,而微分方程求解问题是许多学者研究的热门课题。其中,离散微分方程及其逆问题的求解具有十分重要的意义。本文将支持向量机、Tikhonov正则化和再生核理论相结合,提出一种求解离散线性微分方程及其逆问题的方法。该方法适用于一般的离散线性微分方程及其逆问题的求解,能够得到具有解析表达式的稀疏解,便于后续应用。实验表明,所提出的方法是有效的。

Support vector machines(SVMs) show excellent effectiveness and feasibility in solving approximation problems. Solving differential equations is a hot topic studied by many scholars, in which the solutions of discrete differential equations and their inverse problems are of great significance. A method of solving discrete linear differential equations and their inverse problems is proposed by combining support vector machines, Tikhonov regularization and the theory of reproducing kernels. This method is suitable for solving general discrete linear differential equations and their inverse problems. This method can obtain sparse solutions with analytical expressions, which is convenient for subsequent applications. Experiments show that the proposed method is effective.