DEFIDNET: A framework for optimal allocation of cyberdefenses in Intrusion Detection Networks

Intrusion Detection Networks (IDN) are distributed cyberdefense systems composed of different nodes performing local detection and filtering functions, as well as sharing information with other nodes in the IDN. The security and resilience of such cyberdefense systems are paramount, since an attacker will try to evade them or render them unusable before attacking the end systems. In this paper, we introduce a system model for IDN nodes in terms of their logical components, functions, and communication channels. This allows us to model different IDN node roles (e.g., detectors, filters, aggregators, correlators, etc.) and architectures (e.g., hierarchical, centralized, fully distributed, etc.). We then introduce a threat model that considers adversarial actions executed against particular IDN nodes, and also the propagation of such actions throughout connected nodes. Based on such models, we finally introduce a countermeasure allocation model based on a multi-objective optimization algorithm to obtain optimal allocation strategies that minimize both risk and cost. Our experimental results obtained through simulation with different IDN architectures illustrate the benefit of our framework to design and reconfigure cyberdefense systems optimally.     

盖层封闭能力的灰聚类评价方法

提出灰聚类评价方法，用于盖层封闭能力定量评价。所有被评价对象构成聚类对象集，所有评价指标构成评价指标集，所划分的盖层封闭能力等级构成聚类的灰类集，按照最大聚类权确定对象所属的灰类，即盖层的封闭能力级别。评价流程是：确定指标灰类模式→数据预处理→利用聚类公式计算→利用最大聚类权准则抉择。这种方法避免了加法评分法要依靠人的主观经验划定总评分值的问题，而且由于把评价对象视为“灰”的，具有处理边缘信息的能力。以琼东南盆地３个主要探区的盖层封闭性评价为例，介绍了灰聚类评价的具体应用方法。灰聚类评价方法同样可用于储集层评价、圈闭评价等地质问题的定量研究。表２参３（王孝陵摘     

论环境艺术设计表达教学中范例式教学的重大作用

尽管现代化多种教学手段的多样化发展,范例式教学仍旧以其直观、具体、易理解、易接受,能观察到各个细节等优势,在课堂教学中发挥着重要的作用。近几年,围绕环境艺术设计专业的特点,狠抓以提高教学质量为根本目标的质量工程建设,深化教学方法的改革,特别是加强教学手段的改革,灵活地把范例式教学和多媒体理论教学手段结合起来,提高了教学水平和质量。     

基于自动编码器的深度伪造图像检测方法

基于深度学习的图像伪造方法生成的图像肉眼难辨,一旦该技术被滥用于制作虚假图像和视频,可能会对国家政治、经济、文化造成严重的负面影响,也可能会对社会生活和个人隐私构成威胁。针对上述问题,提出了一种基于自动编码器的深度伪造Deepfake图像检测方法。首先,借助高斯滤波对图像进行预处理,提取高频信息作为模型输入;然后,利用自动编码器对图像进行特征提取,并在编码器中添加注意力机制模块以获取更好的分类效果;最后,通过消融实验证明,采用所提的预处理方法和添加注意力机制模块有助于伪造图像检测。实验结果表明,与ResNet50、Xception以及InceptionV3相比,所提方法在数据集样本量较小且包含的场景丰富时,可以有效检测多种生成方法所伪造的图像,其平均准确率可达97.10%,明显优于对比方法,且其泛化性能也明显优于对比方法。     

考场环境下考生视线估计方法

考生异常行为的监测容易使监考人员产生视觉疲劳。借鉴监考人员发现异常的过程,提出一种可用于考场异常行为分析的视线估计模型。为了减少图像中视线的信息损失,采用注视向量表示视线的大小和方向。该模型分为生成器、视线合成模块、鉴别器,先将考生头部图像输入生成器生成注视向量,再将头部位置和注视位置输入到合成模块得到真实注视向量。将头部图像与上述所得的两种向量输入鉴别器中,其生成对抗模式达到最优时,可得到生成真实值的生成器模型。实验结果表明,在多个考场环境中,该方法的性能优于所对比的几种方法。其中与Lian等人方法相比AUC（Area Under Curve）提高了2.6%,Ang（Angular error）和Dist（Euclidean distance）分别有效降低了20.3%和8.0%。     

页岩气储层常规测井解释模型与应用实例

为深化建南构造侏罗系东岳庙段陆相页岩气储层认识,对东岳庙段岩心与测井资料进行细致研究,建立东岳庙段泥页岩岩性、储层识别、储层矿物含量、孔隙度、渗透率、饱和度和含气量解释计算模型,建立一套较完善的页岩气储层综合解释评价方法。对2口井的东岳庙段页岩气储层进行精细解释,提出东岳庙段页岩气层解释参考标准。适用建南地区东岳庙段页岩气储层测井评价,可以扩展用于整个建南地区及中扬子地区陆相页岩气储层评价。     

规则与实例相结合的公文标题汉英机器翻译系统

随着我国对外开放的不断深入,越来越多的外国朋友希望了解中国的方针政策,而公文是贯彻执行我国方针政策的有力工具,标题作为公文的一大要素,起着点明主题的作用。因此,在信息公开化不断发展的今天,政府网站或公报不仅需要刊载中文版的信息,而且有必要刊载英文版的。鉴于行政机关法定公文标题格式固定、要素稳定的特点,本文基于自建小型公文标题语料库,经分析得出公文标题的表达式和10个词表,然后结合实例,设计一公文标题翻译系统。这只是行政机关公文翻译系统的初步探索,希望能为减轻文秘人员的工作负担、最终实现公文的语言自由转换提供一点帮助。     

Alpha Fusion Adversarial Attack Analysis Using Deep Learning

The deep learning model encompasses a powerful learning ability that integrates the feature extraction, and classification method to improve accuracy. Convolutional Neural Networks (CNN) perform well in machine learning and image processing tasks like segmentation, classification, detection, identification, etc. The CNN models are still sensitive to noise and attack. The smallest change in training images as in an adversarial attack can greatly decrease the accuracy of the CNN model. This paper presents an alpha fusion attack analysis and generates defense against adversarial attacks. The proposed work is divided into three phases: firstly, an MLSTM-based CNN classification model is developed for classifying COVID-CT images. Secondly, an alpha fusion attack is generated to fool the classification model. The alpha fusion attack is tested in the last phase on a modified LSTM-based CNN (CNN-MLSTM) model and other pre-trained models. The results of CNN models show that the accuracy of these models dropped greatly after the alpha-fusion attack. The highest F1 score before the attack was achieved is 97.45 And after the attack lowest F1 score recorded is 22%. Results elucidate the performance in terms of accuracy, precision, F1 score and Recall.     

基于深度卷积生成对抗网络的半生成式视频隐写方案

生成式隐写通过生成足够自然或真实的含密样本来隐藏秘密消息,是信息隐藏方向的研究热点,但目前在视频隐写领域的研究还比较少。结合数字化卡登格的思想,提出一种基于深度卷积生成对抗网络（DCGAN）的半生成式视频隐写方案。该方案中,设计了基于DCGAN的双流视频生成网络,用来生成视频的动态前景、静态后景与时空掩模三个部分,并以随机噪声驱动生成不同的视频。方案中的发送方可设定隐写阈值,在掩模中自适应地生成数字化卡登格,并将其作为隐写与提取的密钥;同时以前景作为载体,实现信息的最优嵌入。实验结果表明,该方案生成的含密视频具有良好的视觉质量,Frechet Inception距离（FID）值为90,且嵌入容量优于现有的生成式隐写方案,最高可达0.11 bpp,能够更高效地传输秘密消息。     

Kernel-based adversarial attacks and defenses on support vector classification

While malicious samples are widely found in many application fields of machine learning, suitable countermeasures have been investigated in the field of adversarial machine learning. Due to the importance and popularity of Support Vector Machines (SVMs), we first describe the evasion attack against SVM classification and then propose a defense strategy in this paper. The evasion attack utilizes the classification surface of SVM to iteratively find the minimal perturbations that mislead the nonlinear classifier. Specially, we propose what is called a vulnerability function to measure the vulnerability of the SVM classifiers. Utilizing this vulnerability function, we put forward an effective defense strategy based on the kernel optimization of SVMs with Gaussian kernel against the evasion attack. Our defense method is verified to be very effective on the benchmark datasets, and the SVM classifier becomes more robust after using our kernel optimization scheme.