Traditional network interdiction refers to the problem of an interdictor trying to reduce the throughput of network users by removing network edges. In this paper, we propose a new paradigm for network interdiction that models scenarios, such as stealth DoS attack, where the interdiction is performed through injecting adversarial traffic flows. Under this paradigm, we first study the deterministic flow interdiction problem, where the interdictor has perfect knowledge of the operation of network users. We show that the problem is highly inapproximable on general networks and is NP-hard even when the network is acyclic. We then propose an algorithm that achieves a logarithmic approximation ratio and quasi-polynomial time complexity for acyclic networks through harnessing the submodularity of the problem. Next, we investigate the robust flow interdiction problem, which adopts the robust optimization framework to capture the case where definitive knowledge of the operation of network users is not available. We design an approximation framework that integrates the aforementioned algorithm, yielding a quasi-polynomial time procedure with poly-logarithmic approximation ratio for the more challenging robust flow interdiction. Finally, we evaluate the performance of the proposed algorithms through simulations, showing that they can be efficiently implemented and yield near-optimal solutions.
ZipNet-GAN: Inferring Fine-grained Mobile Traffic Patterns via a Generative Adversarial Neural Network
大规模移动流量分析对于数字基础设施配置、公共交通、活动规划和其他领域变得至关重要。然而,监控全市范围的移动流量是一个复杂而昂贵的过程,依赖于专用探头。其中一些探测器的精度或覆盖范围有限,另一些每天收集数十GB的日志,这些日志独立提供有限的见解。提取细粒度的模式涉及昂贵的测量、存储和后处理的空间聚合。在本文中,我们提出了一种移动流量超分辨率技术,克服了这些问题,通过推断狭隘的本地化的流量消费从粗糙的测量。我们从图像处理中汲取灵感,设计了一个针对移动网络的深度学习架构,它结合了Zipper网络(ZipNet)和生成对抗神经网络(GAN)模型。这使得能够唯一地捕获在广泛覆盖区域上常规监测的交通量快照(“低分辨率”)与通常在密集计算之后获得的0.05 km $^2$水平上的相应消耗(“高分辨率”)之间的时空关系。我们进行的实验与现实世界的数据集表明,拟议的ZipNet(-GAN)推断流量消耗具有显着的准确性和高达100 $\times$更高的粒度相比,标准的探测,同时优于现有的数据插值技术。据我们所知,这是第一次将超分辨率概念应用于大规模移动交通分析,我们的解决方案是第一个从粗聚合物中推断细粒度城市交通模式的解决方案。
Reliable Feature Selection for Adversarially Robust Cyber-Attack Detection
Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development (GECAD), School of Engineering, Polytechnic of Porto (ISEP/IPP)(波尔图理工学院工程学院智能工程与先进创新计算研究组)
An Adversarial Robustness Benchmark for Enterprise Network Intrusion Detection
Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development (GECAD), School of Engineering, Polytechnic of Porto (ISEP/IPP)(高级创新与发展智能工程与计算研究组(GECAD),波尔图理工学院工程学院(ISEP/IPP))