Semisupervised learning for a hybrid generative/discriminative classifier based on the maximum entropy principle

This paper presents a method for designing semi-supervised classifiers trained on labeled and unlabeled samples. We focus on probabilistic semi-supervised classifier design for multi-class and single-labeled classification problems, and propose a hybrid approach that takes advantage of generative and discriminative approaches. In our approach, we first consider a generative model trained by using labeled samples and introduce a bias correction model, where these models belong to the same model family, but have different parameters. Then, we construct a hybrid classifier by combining these models based on the maximum entropy principle. To enable us to apply our hybrid approach to text classification problems, we employed naive Bayes models as the generative and bias correction models. Our experimental results for four text data sets confirmed that the generalization ability of our hybrid classifier was much improved by using a large number of unlabeled samples for training when there were too few labeled samples to obtain good performance. We also confirmed that our hybrid approach significantly outperformed generative and discriminative approaches when the performance of the generative and discriminative approaches was comparable. Moreover, we examined the performance of our hybrid classifier when the labeled and unlabeled data distributions were different.     

Adaptive semi-supervised learning on labeled and unlabeled data with different distributions

Developing methods for designing good classifiers from labeled samples whose distribution is different from that of test samples is an important and challenging research issue in the fields of machine learning and its application. This paper focuses on designing semi-supervised classifiers with a high generalization ability by using unlabeled samples drawn by the same distribution as the test samples and presents a semi-supervised learning method based on a hybrid discriminative and generative model. Although JESS-CM is one of the most successful semi-supervised classifier design frameworks based on a hybrid approach, it has an overfitting problem in the task setting that we consider in this paper. We propose an objective function that utilizes both labeled and unlabeled samples for the discriminative training of hybrid classifiers and then expect the objective function to mitigate the overfitting problem. We show the effect of the objective function by theoretical analysis and empirical evaluation. Our experimental results for text classification using four typical benchmark test collections confirmed that with our task setting in most cases, the proposed method outperformed the JESS-CM framework. We also confirmed experimentally that the proposed method was useful for obtaining better performance when classifying data samples into either known or unknown classes, which were included in given labeled samples or not, respectively.     

基于条件对抗域泛化的人脸活体检测方法

针对现存的跨场景人脸活体检测模型泛化性能差、类间重叠等问题,提出了一种基于条件对抗域泛化的人脸活体检测方法。首先,该方法使用嵌入注意力机制的U-Net网络和ResNet-18编码器提取多个源域的特征,然后将提取的特征送入辅助分类器,并将特征编码器的输出和分类器预测的结果通过多线性映射的方法进行融合,再输入到域判别器中进行对抗训练,以实现特征和类层面对齐多个源域。其次,为了减少预测不准确的难迁移样本对域泛化造成的影响,采用了熵函数来控制样本的优先级,以提高域泛化的性能。此外,通过添加人脸深度图以进一步抓取活体与假体的区别特征,通过非对称三元组损失约束作为辅助监督,进一步提高类内紧凑性和类间区分性。在公开活体检测数据集上的对比实验验证了所提方法的有效性。     

面向网络流量入侵检测系统的黑盒攻击

基于机器学习的网络流量检测系统是网络安全领域现阶段比较热门的研究方向,但同时网络流量检测系统又受到了巨大挑战,因为攻击样本的生成,使该检测系统对恶意流量的检测性能降低.使用生成对抗网络生成对抗样本,通过在原始恶意流量中加入噪声干扰,即在攻击特征中加入不影响原始流量特性的非定向扰动,来实现扰乱检测模型的判断,从而躲过特征...     

An efficient cascaded method for network intrusion detection based on extreme learning machines

Machine learning techniques are widely used for network intrusion detection (NID). However, it has to face the unbalance of training samples between classes as it is hard to collect samples of some intrusion classes. This would produce false positives for these intrusion classes. Meanwhile, since there are various types of intrusions, classification boundaries between different classes are seriously nonlinear. Due to the huge amount of training data, computational efficiency is also required. This paper therefore proposes an efficient cascaded classifier for NID. This classifier consists of a collection of binary base classifiers which are serially connected. Each base classifier corresponds to a type of intrusion. The order of these base classifiers is automatically determined based on the number of false positives to cope with the unbalance of training samples. Extreme learning machine algorithm, which has low computational cost, is used to train these base classifiers to delineate the nonlinear boundaries between classes. This proposed NID method is evaluated on the KDD99 data set. Experimental results have shown that this proposed method outperforms other state-of-the-art methods including decision tree, back-propagation neural network and support vector machines.     

对抗学习遥感图像场景识别

目的 在高分辨率遥感图像场景识别问题中,经典的监督机器学习算法大多需要充足的标记样本训练模型,而获取遥感图像的标注费时费力。为解决遥感图像场景识别中标记样本缺乏且不同数据集无法共享标记样本问题,提出一种结合对抗学习与变分自动编码机的迁移学习网络。方法 利用变分自动编码机（variational auto-encoders,VAE）在源域数据集上进行训练,分别获得编码器和分类器网络参数,并用源域编码器网络参数初始化目标域编码器。采用对抗学习的思想,引入判别网络,交替训练并更新目标域编码器与判别网络参数,使目标域与源域编码器提取的特征尽量相似,从而实现遥感图像源域到目标域的特征迁移。结果 利用两个遥感场景识别数据集进行实验,验证特征迁移算法的有效性,同时尝试利用SUN397自然场景数据集与遥感场景间的迁移识别,采用相关性对齐以及均衡分布适应两种迁移学习方法作为对比。两组遥感场景数据集间的实验中,相比于仅利用源域样本训练的网络,经过迁移学习后的网络场景识别精度提升约10%,利用少量目标域标记样本后提升更为明显;与对照实验结果相比,利用少量目标域标记样本时提出方法的识别精度提升均在3%之上,仅利用源域标记样本时提出方法场景识别精度提升了10%~40%;利用自然场景数据集时,方法仍能在一定程度上提升场景识别精度。结论 本文提出的对抗迁移学习网络可以在目标域样本缺乏的条件下,充分利用其他数据集中的样本信息,实现不同场景图像数据集间的特征迁移及场景识别,有效提升遥感图像的场景识别精度。     

基于多尺度特征多对抗网络的雾天图像识别

目的 当前的大型数据集,例如ImageNet,以及一些主流的网络模型,如ResNet等能直接高效地应用于正常场景的分类,但在雾天场景下则会出现较大的精度损失。雾天场景复杂多样,大量标注雾天数据成本过高,在现有条件下,高效地利用大量已有场景的标注数据和网络模型完成雾天场景下的分类识别任务至关重要。方法 本文使用了一种低成本的数据增强方法,有效减小图像在像素域上的差异。基于特征多样性和特征对抗的思想,提出多尺度特征多对抗网络,通过提取数据的多尺度特征,增强特征在特征域分布的代表性,利用对抗机制,在多个特征上减少特征域上的分布差异。通过缩小像素域和特征域分布差异,进一步减小领域偏移,提升雾天场景的分类识别精度。结果 在真实的多样性雾天场景数据上,通过消融实验,使用像素域数据增强方法后,带有标签的清晰图像数据在风格上更趋向于带雾图像,总的分类精度提升了8.2%,相比其他的数据增强方法,至少提升了6.3%,同时在特征域上使用多尺度特征多对抗网络,相比其他的网络,准确率至少提升了8.0%。结论 像素域数据增强以及多尺度特征多对抗网络结合的雾天图像识别方法,综合考虑了像素域和特征域的领域分布差异,结合了多尺度的丰富特征信息,同时使用多对抗来缩小雾天数据的领域偏移,在真实多样性雾天数据集上获得了更好的图像分类识别效果。     

Multi-mode data augmentation and fault diagnosis of rotating machinery using modified ACGAN designed with new framework

As one of the representative unsupervised data augmentation methods, generative adversarial networks (GANs) have the potential to solve the problem of insufficient samples in fault diagnosis of rotating machinery. However, the existing unsupervised GANs are usually incapable of simultaneously generating multi-mode fault samples and have some shortcomings such as mode collapse and gradient vanishing. To overcome these deficiencies, a supervised model called modified auxiliary classifier GAN (MACGAN) designed with new framework is proposed in this paper. Firstly, a new ACGAN framework is developed by adding an independent classifier to improve the compatibility between the classification and discrimination. Secondly, the Wasserstein distance is introduced in the new loss functions to overcome mode collapse and gradient vanishing. Finally, to achieve stable training, a spectral normalization is used to replace the weight clipping to constrain the weight parameters of discriminator. The proposed method is applied to fault diagnosis of bearing and gear. Compared with the existing GANs, the proposed method can more efficiently generate multi-mode fault samples with higher qualities, which can be used to assist the training of deep learning-based fault diagnosis models with high accuracy and good stability.     

基于CNN的改进行人重识别技术

针对行人重识别研究中训练样本的不足,为提高识别精度及泛化能力,提出一种基于卷积神经网络的改进行人重识别方法。首先对训练数据集进行扩充,使用生成对抗网络无监督学习方法生成无标签图像;然后与原数据集联合作半监督卷积神经网络训练,通过构建一个Siamese网络,结合分类模型和验证模型的特点进行训练;最后加入无标签图像类别分布方法,计算交叉熵损失来进行相似度量。实验结果表明,在Market-1501、CUHK03和DukeMTMC-reID数据集上,该方法相比原有的Siamese方法在Rank-1和mAP等性能指标上有近3～5个百分点的提升。当样本较少时,该方法具有一定应用价值。     

采用双框架生成对抗网络的图像运动模糊盲去除

传统运动模糊盲去除方法需先预测模糊图像的模糊核,再复原清晰图像。而实际 环境中的复杂的模糊核使此方法不能在视觉上很好地减小实际图像和复原后图像的差异,且直 接将现流行的生成对抗模型应用在图像模糊盲去除任务中会有严重的模式崩塌现象。因此,围 绕去模糊任务的特点提出了一种端到端的生成对抗网络模型--双框架生成对抗网络。该方案 不需要预测模糊核,直接实现图片运动模糊的盲去除。双框架生成对抗网络在原有 CycleGan 基础上将其网络结构和损失函数均作出了改进,提高了运动图像盲去除的精度,并且在样本有 限情况下大幅度增强了网络的稳定性。实验采用最小均方差优化网络训练,最后通过生成网络 和判别网络对抗训练获得清晰图像。在 ILSVRC2015 VID 数据集上的实验结果表明,该方法复原 质量更高,且复原结果在后续目标检测任务中达到了更优的效果。     

Towards Robust Adversarial Training via Dual-label Supervised and Geometry Constraint

Recent studies have shown that adversarial training is an effective method to defend against adversarial sample attacks. However, existing adversarial training strategies improve the model robustness at a price of a lowered generalization ability of the model. At this stage, the mainstream adversarial training methods usually deal with each training sample independently and ignore the inter-sample relationships, which prevents the model from fully exploiting the geometric relationship between samples to learn a more robust model for better defense against adversarial attacks. Therefore, this paper focuses on how to maintain the stability of the geometric structure between samples during adversarial training to improve the model robustness. Specifically, in adversarial training, a new geometric structure constraint method is designed with the aim to maintain the consistency of the feature space distribution between normal samples and adversarial samples. Furthermore, a dual-label supervised learning method is proposed, which leverages the labels of both natural samples and adversarial samples for joint supervised training of the model. Lastly, the characteristics of the dual-label supervised learning method are analyzed, and the working mechanism of the adversarial samples are explained theoretically. It is concluded from extensive experiments on benchmark datasets that the proposed approach effectively improves the robustness of the model while maintaining good generalization accuracy. The related code has been open-sourced: https://github.com/SkyKuang/DGCAT     

基于梯形网络和改进三训练法的半监督分类

为了提高半监督深层生成模型的分类性能, 提出一种基于梯形网络和改进三训练法的半监督分类模型. 该模型在梯形网络框架有噪编码器的最高层添加3个分类器, 结合改进的三训练法提高图像分类性能. 首先, 用基于类别抽样的方法将有标记数据分为3份, 模型以有标记数据的标签误差和未标记数据的重构误差相结合的方式调整参数, 训练得到3个Large-margin Softmax分类器; 接着, 用改进的三训练法对未标记数据添加伪标签, 并对新的标记数据分配不同权重, 扩充训练集; 最后, 利用扩充的训练集更新模型. 训练完成后, 对分类器进行加权投票, 得到分类结果. 模型得到的梯形网络的特征有更好的低维流形表示, 可以有效地避免因为样本数据分布不均而导致的分类误差, 增强泛化能力. 模型分别在MNIST数据库, SVHN数据库和CIFAR10数据库上进行实验, 并且与其他半监督深层生成模型进行了比较, 结果表明本文所提出的模型得到了更高的分类精度.     

一种协同半监督分类算法Co-S3OM

为了提高半监督分类的有效性, 提出了一种基于SOM神经网络和协同训练的半监督分类算法Co-S3OM (coordination semi-supervised SOM)。将有限的有标记样本分为无重复的三个均等的训练集, 分别使用改进的监督SSOM算法(supervised SOM)训练三个单分类器, 通过三个单分类器共同投票的方法挖掘未标记样本中的隐含信息, 扩大有标记样本的数量, 依次扩充单分类器训练集, 生成最终的分类器。最后选取UCI数据集进行实验, 结果表明Co-S3OM具有较高的标记率和分类率。     

基于自编码器的网络异常检测研究综述

网络入侵检测技术是指对危害计算机系统安全的行为进行检测的方法,它是计算机网络安全领域中的必不可少的防御机制。目前,基于有监督学习的网络异常入侵检测技术具有较高的效率和准确率,该类方法获得了广泛关注,取得了大量的研究成果。但是这类方法需要借助大量标注样本进行模型训练。为减少对标注样本依赖,基于无监督学习或半监督学习的网络入侵检测技术被提出,并逐渐成为该领域的研究热点。其中,基于自编码器的网络异常检测技术是这方面技术的典型代表。该文首先介绍了各类自编码器的基本原理、模型结构、损失函数和训练方法。然后在此基础上将其分为基于阈值和基于分类的方法。其中,基于阈值的方法用又可分为基于重构误差和基于重构概率两类。合适的阈值对异常检测技术的成败至关重要,该文介绍了三种阈值的计算方法。接着对比分析了多个代表性研究工作的方法、性能及创新点,最后对该研究中存在的问题做了介绍,并对未来的研究方向做了展望。     

Defeating traffic analysis via differential privacy: a case study on streaming traffic

In this paper, we explore the adaption of techniques previously used in the domains of adversarial machine learning and differential privacy to mitigate the ML-powered analysis of streaming traffic. Our findings are twofold. First, constructing adversarial samples effectively confounds an adversary with a predetermined classifier but is less effective when the adversary can adapt to the defense by using alternative classifiers or training the classifier with adversarial samples. Second, differential-privacy guarantees are very effective against such statistical-inference-based traffic analysis, while remaining agnostic to the machine learning classifiers used by the adversary. We propose three mechanisms for enforcing differential privacy for encrypted streaming traffic and evaluate their security and utility. Our empirical implementation and evaluation suggest that the proposed statistical privacy approaches are promising solutions in the underlying scenarios

     

Multi-level hybrid support vector machine and extreme learning machine based on modified K-means for intrusion detection system

Intrusion detection has become essential to network security because of the increasing connectivity between computers. Several intrusion detection systems have been developed to protect networks using different statistical methods and machine learning techniques. This study aims to design a model that deals with real intrusion detection problems in data analysis and classify network data into normal and abnormal behaviors. This study proposes a multi-level hybrid intrusion detection model that uses support vector machine and extreme learning machine to improve the efficiency of detecting known and unknown attacks. A modified K-means algorithm is also proposed to build a high-quality training dataset that contributes significantly to improving the performance of classifiers. The modified K-means is used to build new small training datasets representing the entire original training dataset, significantly reduce the training time of classifiers, and improve the performance of intrusion detection system. The popular KDD Cup 1999 dataset is used to evaluate the proposed model. Compared with other methods based on the same dataset, the proposed model shows high efficiency in attack detection, and its accuracy (95.75%) is the best performance thus far.     

基于聚类簇结构特性的自适应综合采样法在入侵检测中的应用

基于机器学习的网络入侵检测方法将恶意网络行为(入侵)检测转化为模式识别(分类)问题,因其适应性强、灵敏度高等优点,受到国内外广泛关注.然而,现有的模式分类器往往假设数据集的分布是均衡的,而真实的网络环境中,入侵行为要远少于正常访问,这给网络入侵行为检测带来巨大挑战.因此,提出一种基于聚类簇结构特性的综合采样法(CSbADASYN),通过挖掘少数类样本的内部结构对其进行自适应过采样,以获得样本分布结构特性保持的均衡数据样本,解决因数据不均衡带来的分类偏向.CSbADASYN先采用谱聚类方法对数据集中的少数类样本进行聚类分析,再根据所获得的聚类簇结构自适应插值,将获得样本分布结构保持的均衡样本用于分类器模型学习.在经典的NSL-KDD和KDD99数据集上进行大量的验证性和对比性实验,结果表明,CSbADASYN 能使传统分类器模型在不均衡数据集上的分类性能得到明显提升.与传统的未经样本均衡处理和其他的带均衡处理的入侵检测方法相比,该方法能获得更低的误报率和漏报率.     

A weighted intrusion detection model of dynamic selection

In view of the difficulty of existing intrusion detection methods in dealing with new forms, large scale, and high concealment of network intrusion behaviors, this paper presents a weighted intrusion detection model of the dynamic selection (WIDMoDS) based on data features. The aim is to customize intrusion detection models for network intrusion data sets of different types, sizes and structures. First, according to data features, single classifiers are clustered using a hierarchical clustering algorithm based on the classifiers evaluation indicators, and then, the classifiers selection is by means of accuracy of the single classifiers, in addition, the data-classifier applicable indicators (DCAI) and of the classifiers performances are used for calculating the weights of subjective and objective, and then calculating combined weight ranks. Finally, a custom intrusion detection model is generated by the Weight-voting (W-voting) algorithm. Our experiments show that this model can optimize the number of classifiers based on the data sets features, reduce the problem of redundant or insufficient classifiers in the ensemble process. A new network intrusion detection model of combining the classifier characteristics with the dataset attributes can improve the accuracy of intrusion detection.

     

Neural network-based symbol recognition using a few labeled samples

The recognition of pen-based visual patterns such as sketched symbols is amenable to supervised machine learning models such as neural networks. However, a sizable, labeled training corpus is often required to learn the high variations of freehand sketches. To circumvent the costs associated with creating a large training corpus, improve the recognition accuracy with only a limited amount of training samples and accelerate the development of sketch recognition system for novel sketch domains, we present a neural network training protocol that consists of three steps. First, a large pool of unlabeled, synthetic samples are generated from a small set of existing, labeled training samples. Then, a Deep Belief Network (DBN) is pre-trained with those synthetic, unlabeled samples. Finally, the pre-trained DBN is fine-tuned using the limited amount of labeled samples for classification. The training protocol is evaluated against supervised baseline approaches such as the nearest neighbor classifier and the neural network classifier. The benchmark data sets used are partitioned such that there are only a few labeled samples for training, yet a large number of labeled test cases featuring rich variations. Results suggest that our training protocol leads to a significant error reduction compared to the baseline approaches.     

主动协同半监督粗糙集分类模型

粗糙集理论是一种有监督学习模型,一般需要适量有标记的数据来训练分类器。但现实一些问题往往存在大量无标记的数据,而有标记数据由于标记代价过大较为稀少。文中结合主动学习和协同训练理论,提出一种可有效利用无标记数据提升分类性能的半监督粗糙集模型。该模型利用半监督属性约简算法提取两个差异性较大的约简构造基分类器,然后基于主动学习思想在无标记数据中选择两分类器分歧较大的样本进行人工标注,并将更新后的分类器交互协同学习。UCI数据集实验对比分析表明,该模型能明显提高分类学习性能,甚至能达到数据集的最优值。