摘要
近年来基于深度学习的AI技术发展迅速,数据规模成为需要考虑的首要条件。如何通过已有小样本数据集实现数据增强成为一个值得关注的问题。针对上述问题对数据增强方法进行概括总结,介绍目前较为主流的数据生成模型-生成对抗网络(GAN),着重对基于GAN的数据增强方法在网络安全领域小规模数据集上的研究应用进行分析总结,具体涵盖方法所用的模型结构、创新之处、数据集、评估结果,提出对其的展望。
In recent years,AI technology based on deep learning has developed rapidly,the scale of data is the first condition that deep learning needs to consider.How to implement data enhancement with existing small sample data sets has become a problem worthy of attention.In order to solve the above problems,we summarized the data enhancement methods,introduced the current mainstream data generation model-generative adversarial network(GAN),and focused on the research and application of GAN-based data enhancement methods on small-scale data sets in the field of network security.We analyzed and summarized the model structure,innovations,data sets and evaluation results used in the method,and put forward the outlook for it.
作者
朱晓慧
钱丽萍
傅伟
Zhu Xiaohui;Qian Liping;Fu Wei(School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 100044,China;Beijing Key Laboratory of Intelligent Processing for Building Big Data,Beijing 100044,China)
出处
《计算机应用与软件》
北大核心
2022年第11期288-296,共9页
Computer Applications and Software
基金
国家自然科学基金项目(61571144)。
关键词
深度学习
生成对抗网络
数据增强
小样本数据
网络安全
Deep learning
Generative adversarial network
Data enhancement
Small sample data
Cyber security