交互式遗传算法中用户评估质量和收敛性能提高

AN IMPROVEMENT OF CONVERGENCE AND USER EVALUATION QUALITY FOR INTERACTIVE GENETIC ALGORITHMS

交互式遗传算法主要是针对一些应用领域存在的适应度函数难以明确表达的问题,通过用户参与遗传操作过程,对进化个体进行评估以代替计算过程的一种改进方法。但交互式遗传算法存在的主要问题是:当遗传操作的收敛速度慢时,用户需对大量个体进行评估,尤其是在个体间相似程度较高时,容易产生疲劳现象。针对这种问题,本文提出三种方法来提高收敛性及评估过程的质量。在人脸图形上所做的实验结果表明,算法可以有效地缓解用户的疲劳问题。

Interactive GA is an improved method which is well suited to solve the problem with regard to the fitness function being hard to express explicitly. In such a way,evaluation of individuals is easy for users to perform. However,an existing significant problem for IGA is that users have to evaluate a large number of individuals when the convergence speed of genetic operations is too slow. Furthermore, users tend to be tired especially when the similarity among some individuals is relatively high.To deal with this problem, three approaches are proposed to improve the convergence and the quality of evaluation. It has been shown that the problem of tiredness for users can effectively be solved from the experimental results obtained in the human face graphics.