说明：  进化编程可以通过演化数字向量的种群来解决黑盒函数优化问题。 进化过程中的变化分量由变异算子提供，该变换器通常是高斯，柯西或列维概率分布。 在本文中，我们使用遗传编程自动生成进化程序系统的变异算子，在一组函数类上测试所提出的方法，这些函数类代表了函数的来源。一组基准函数类的实证结果说明了遗传编程的canevolve 变异运算符，从训练集到每个函数类的测试集很好地推广。 在大多数情况下，所提出的方法能够胜过具有统计学意义的现有人类设计的变异算子，其余部分观察到竞争结果。
(In this paper, we usegenetic programming to automatically generate mutation operators for an evolutionary programmingsystem, testing the proposed approach over a set of function classes, which represent a source of functions.The empirical results over a set of benchmark function classes illustrate that genetic programming canevolve mutation operators which generalise well from the training set to the test set on each functionclass. The proposed method is able to outperform existing human designed mutation operators withstatistical significance in most cases, with competitive results observed for the rest.)