当前位置: X-MOL 学术J. Appl. Stat. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Modeling multivariate cyber risks: deep learning dating extreme value theory
Journal of Applied Statistics ( IF 1.2 ) Pub Date : 2021-06-04 , DOI: 10.1080/02664763.2021.1936468
Mingyue Zhang Wu 1 , Jinzhu Luo 2 , Xing Fang 2 , Maochao Xu 3 , Peng Zhao 1
Affiliation  

Modeling cyber risks has been an important but challenging task in the domain of cyber security, which is mainly caused by the high dimensionality and heavy tails of risk patterns. Those obstacles have hindered the development of statistical modeling of the multivariate cyber risks. In this work, we propose a novel approach for modeling the multivariate cyber risks which relies on the deep learning and extreme value theory. The proposed model not only enjoys the high accurate point predictions via deep learning but also can provide the satisfactory high quantile predictions via extreme value theory. Both the simulation and empirical studies show that the proposed approach can model the multivariate cyber risks very well and provide satisfactory prediction performances.



中文翻译:

多元网络风险建模:深度学习约会极值理论

网络风险建模一直是网络安全领域的一项重要但具有挑战性的任务,这主要是由风险模式的高维度和重尾造成的。这些障碍阻碍了多元网络风险统计模型的发展。在这项工作中,我们提出了一种基于深度学习和极值理论的多变量网络风险建模新方法。所提出的模型不仅可以通过深度学习进行高精度点预测,还可以通过极值理论提供令人满意的高分位数预测。仿真和实证研究均表明,所提出的方法可以很好地模拟多变量网络风险,并提供令人满意的预测性能。

更新日期:2021-06-04
down
wechat
bug