A fundamental study on adaptive surrogate-assisted evolutionary computation using rank correlation
A fundamental study on adaptive surrogate-assisted evolutionary computation using rank correlation
复制标题
DOI：
10.1109/snpd.2017.8022789
复制
发表时间：
2017-06
期刊：
影响因子：
--
通讯作者：
Yudai Kuwahata;J. Kushida;S. Ono
中科院分区：
文献类型：
--
作者：
Yudai Kuwahata;J. Kushida;S. Ono
代理辅助进化计算（SAEC）已广泛应用于目标函数的逼近。然而，SAEC也可能潜在地减少廉价的优化问题的处理时间，其中在几秒或几分钟内评估解决方案。为了实现这一点，适应度函数的近似模型应该在优化期间尽可能少地迭代。本文提出了一种自适应SAEC算法，利用目标函数的实际评估值和近似评估值之间的秩相关性。然后，这些相关性用于自适应地切换近似和实际评估阶段，减少学习近似模型所需的运行次数。实验结果表明，该方法可以成功地减少一些基准函数的处理时间，即使在廉价的情况下。
Surrogate-Assisted Evolutionary Computation (SAEC) has widely applied to approximate an objective function. However, SAEC may potentially also reduce the processing time of inexpensive optimization problems wherein solutions are evaluated within a few seconds or minutes. To achieve this, the approximation model of a fitness function should be iterated as few times as possible during optimization. This paper proposes an adaptive SAEC algorithm using the rank correlations between the actually evaluated and approximately evaluated values of the objective function. These correlations are then used to adaptively switch the approximation and actual evaluation phases, reducing the number of runs required to learn the approximation model. It was confirmed experimentally that the proposed method could successfully reduce the processing time in some benchmark functions even under inexpensive scenario.