支持向量回归机元参数优化方法

为了优化ε不敏感支持向量回归机（ε-support vector regression, ε-SVR）的三类元参数，根据其耦合程度将其优化问题分解为核参数优化和结构参数（即不敏感参数和正则化参数）优化两个子问题，并提出了相应的优化方法。首先，提出了一种新的核校准系数以优化核参数；其次，提出了一种基于期望训练误差的结构参数优化方法；最后，为准确估算ε-SVR的期望训练误差，还提出了一种根据实际训练误差分布特征评估和校正期望误差的方法。仿真结果表明，该文方法具有与交叉检验法近似的优化效果，且时间效率更高。

Meta-parameters optimization method for support vector regression

To optimize the meta parameters of ε insensitive support vector regression (ε-SVR), the meta parameters optimization problem is divided into two sub-problems named as kernel parameter optimization and structure parameters (including insensitive parameter and regularization parameter) optimization according to the coupling degrees among them, and corresponding optimization methods are proposed. First, a new kernel  alignment coefficient is proposed for the former. Second, a method based on expectation training error is proposed for the latter. Finally, to estimate accurately the expectation error of the ε-SVR, a method to evaluate and adjust expectation error according to the distribution characteristics of the real training errors is proposed. Simulation results show that the proposed method is nearly as accurate as the cross validation method, and much more rapid.