Abstract

In this paper, a novel and powerful metaheuristic optimizer, named the growth optimizer (GO), is proposed. Its main design inspiration originates from the learning and reflection mechanisms of individuals in their growth processes in society. Learning is the process of individuals growing up by acquiring knowledge from the outside world. Reflection is the process of checking the individual’s own deficiencies and adjusting the individual’s learning strategies to help the individual’s growth.

本文提出一种新颖而强大的元启发式优化器——生长优化器（GO）。其设计灵感源于个体在社会成长过程中的学习与反思机制：学习是个体从外部获取知识、促进自身成长的过程；反思是审视自身不足、调整学习策略以助力成长的过程。本文将这种成长行为数学化，并在 CEC 2017 基准的 30 个国际测试函数上与 50 个先进元启发式算法对比。收敛精度对比及 Friedman 检验与 Wilcoxon 符号秩检验表明 GO 具有竞争力。在基于隐马尔可夫模型的多序列比对与基于 Kapur 熵的多阈值图像分割两类实际问题上，GO 也取得更优结果。

Keywords