进化算法在复杂网络中的应用综述

Survey on Application of Evolutionary Algorithms in Complex Networks

复杂网络作为现今科学研究中的一个热点学科,在过去20年里得到了巨大的发展.现实中大量的复杂的交互系统,比如互联网、交通运输网、神经网络等都可以抽象为复杂网络,以进行系统的分析和研究.进化算法作为优化工具应用于复杂网络的不同领域的各个任务中,如网络社团结构的检测任务、网络动力学中的鲁棒性优化任务、网络传播中关键节点的搜寻任务等.本文首先对复杂网络和进化算法相关的基础知识进行了全面的概述,重点讨论了复杂网络中目标优化的研究进展,针对不同任务对优化目标及其具体应用展开了详细介绍,同时,对算法的性能评价指标进行了概述.此外,本文通过一系列实验展示了单/多目标优化算法在复杂网络优化问题上的性能表现,以及部分目标之间的相关性关系.最后对复杂网络中优化问题未来的研究动向进行了展望,为今后研究人员开展进化计算和复杂网络相结合的相关研究提供一些思路.

As a hot subject in current scientific research,complex networks have developed tremendously in the past 20 years.A large number of complex interactive systems in reality,such as the Internet,transportation networks,and neural networks,can be abstracted as complex networks for systematic analysis and research.As an optimization tool,evolutionary algorithms are applied to various tasks in different fields of complex networks,such as the tasks of detecting network community structure,optimizing robustness in network dynamics,and searching key nodes in network propagation.This study first gives a comprehensive overview of the basic knowledge related to complex networks and evolutionary algorithms and then focuses on the research progress of objective optimization in complex networks.The optimization objectives and their specific applications are introduced in detail for different tasks.At the same time,the performance evaluation indexes of the algorithms are summarized.In addition,this study demonstrates the performance of single/multi-objective optimization algorithms on complex network optimization problems through a series of experiments and represents the correlation between some objectives.In the last part of the study,the future research trends of optimization problems in complex networks are prospected,which can provide some ideas for researchers to carry out related research on the combination of evolutionary computation and complex networks.