代理模型辅助的复杂网络能控性鲁棒性优化方法  

Controllability Robustness Optimization Method of Complex Networks Based on Surrogate Model Assistance

作　　者：聂君凤 于卓然 李均利[1] NIE Junfeng;YU Zhuoran;LI Junli(School of Computer Science,Sichuan Normal University,Chengdu 610101,China)

机构地区：[1]四川师范大学计算机科学学院,成都610101

出　　处：《小型微型计算机系统》2024年第1期151-159,共9页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(62002249)资助.

摘　　要：近年来,复杂网络的鲁棒性优化问题引起人们广泛关注.复杂网络暴露在外会受到各种各样的攻击,因此如何设计抗击能力较好的网络结构成为了研究热点.虽然现有的方法在小规模复杂网络的鲁棒性方面已经取得了显著成果,但大规模复杂网络的能控性鲁棒性优化的计算成本非常大.而代理模型可以以较低的计算成本来代替优化过程中对复杂网络能控性鲁棒性的评估,但一个代理模型不可能适用于评估所有类型的复杂网络能控性鲁棒性.文中将Dempster-Shafer理论应用于代理模型选择及其混合,并把选择出的代理模型用来辅助进化算法搜索能控性鲁棒性更优的网络结构.此方法在SF、ER、SW、RR、RT和QS 6种合成网络上的实验结果表明:在不同类型的复杂网络中选择合适的代理模型能更好的辅助进化算法找到能控性鲁棒性更优的网络结构.In recent years,the robust optimization of complex networks has attracted extensive attention.Complex networks are exposed to a variety of attacks,so how to design a network structure with good resistance has become a research hotspot.Although the existing methods have achieved remarkable results in the robustness of small-scale complex networks,the computational cost of controllability robustness optimization of large-scale complex networks is very large.The surrogate model can replace the evaluation of controllability robustness of complex networks in the optimization process with a lower computational cost,but a surrogate model can′t be used to evaluate the controllability robustness of all types of complex networks.In this paper,Dempster-Shafer theory is applied to surrogate model selection and its mixture,and the selected surrogate model is used to assist evolutionary algorithm to search for a network structure with better controllability and robustness.The experimental results on six synthetic networks of SF,ER,SW,RR,RT and QS show that selecting the appropriate surrogate model in different types of complex networks can better assist evolutionary algorithm to find the network structure with better controllability robustness.

关 键 词：代理模型 进化算法 复杂网络 能控性鲁棒性 D-S理论 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]