车联网中基于移动边缘计算的内容感知分类卸载算法研究

随着智能交通的快速发展,车辆终端产生大量需要实时处理的数据消息,而在有限资源上的竞争将会增加消息处理的时延,且对终端设备造成很大的能量消耗。针对时延和能量损耗的均衡关系,该文提出一种基于移动边缘计算(MEC)的内容感知分类卸载算法。首先根据层次分析法对安全消息进行优先级划分,然后建立时延和能量损耗的最优任务卸载模型,通过给时延和能量损耗赋予不同的权重系数构造关系模型,并利用拉格朗日松弛法将非凸问题转化为凸问题,从而结合次梯度投影法和贪婪算法得到问题的可行解。性能评估结果表明,该算法在一定程度上改善了消息处理时延和能量损耗。     

最小化类内距离和分类算法

支持向量机分类算法引入惩罚因子来调节过拟合和线性不可分时无解的问题,优点是可以通过调节参数取得最优解,但带来的问题是允许一部分样本错分。错分的样本在分类间隔之间失去了约束,导致两类交界处样本杂乱分布,并且增加了训练的负担。为了解决上述问题,该文根据大间隔分类思想,基于类内紧密类间松散的原则,提出一种新的分类算法,称之为最小化类内距离和(Intraclass-Distance-Sum-Minimization, IDSM)分类算法。该算法根据最小化类内距离和准则构造训练模型,通过解析法求解得到最佳的映射法则,进而利用该最佳映射法则对样本进行投影变换以达到类内间隔小类间间隔大的效果。相应地,为解决高维样本分类问题,进一步提出了该文算法的核化版本。在大量UCI数据集和Yale大学人脸数据库上的实验结果表明了该文算法的优越性。     

Robust and Efficient Pattern Classification using Large Geometric Margin Minimum Classification Error Training

Recently, one of the standard discriminative training methods for pattern classifier design, i.e., Minimum Classification Error (MCE) training, has been revised, and its new version is called Large Geometric Margin Minimum Classification Error (LGM-MCE) training. It is formulated by replacing a conventional misclassification measure, which is equivalent to the so-called functional margin, with a geometric margin that represents the geometric distance between an estimated class boundary and its closest training pattern sample. It seeks the status of the trainable classifier parameters that simultaneously correspond to the minimum of the empirical average classification error count loss and the maximum of the geometric margin. Experimental evaluations showed the fundamental utility of LGM-MCE training. However, to increase its effectiveness, this new training required careful setting for hyperparameters, especially the smoothness degree of the smooth classification error count loss. Exploring the smoothness degree usually requires many trial-and-error repetitions of training and testing, and such burdensome repetition does not necessarily lead to an optimal smoothness setting. To alleviate this problem and further increase the effect of geometric margin employment, we apply in this paper a new idea that automatically determines the loss smoothness of LGM-MCE training. We first introduce a new formalization of it using the Parzen estimation of error count risk and formalize LGM-MCE training that incorporates a mechanism of automatic loss smoothness determination. Importantly, the geometric-margin-based misclassification measure adopted in LGM-MCE training is directly linked with the geometric margin in a pattern sample space. Based on this relation, we also prove that loss smoothness affects the production of virtual samples along the estimated class boundaries in pattern sample space. Finally, through experimental evaluations and in comparisons with other training methods, we elaborate the characteristics of LGM-MCE training and its new function that automatically determines an appropriate loss smoothness degree.     

基于密度敏感最大软间隔SVDD不均衡数据分类算法

为了提高传统支持向量域描述（C-SVDD）算法处理不均衡数据集的分类能力,提出一种基于密度敏感最大软间隔支持向量域描述（DSMSM-SVDD）算法.该算法通过对多数类样本引入相对密度来体现训练样本原始空间分布对求解最优分类界面的影响,通过在目标函数中增加最大软间隔正则项,使C-SVDD的分类边界向少数类偏移,进而提高算法分类性能.算法首先对每个多数类样本计算相对密度来反映样本的重要性,然后将训练样本输入到DSMSM-SVDD中实现数据分类.实验部分,讨论了算法参数间的关系及其对算法分类性能的影响,给出算法参数取值建议.最后通过与C-SVDD的对比实验,表明本文建议的算法在不均衡数据情况下的分类性能优于C-SVDD算法.     

A weakly supervised large margin domain adaptation method for isolated handwritten digit recognition

Learning handwriting categories fail to perform well when trained and tested on data from different databases. In this paper, we propose a novel large margin domain adaptation algorithm which is able to learn a transformation between training and test datasets in addition to adapting the parameters of classifier using a few or even no training labeled samples from target handwriting dataset. Additionally, we developed a framework of ensemble projection feature learning for datasets representation as a front end for our algorithm to utilize the abundant unlabeled samples in target domain. Experiments on different handwritten digit datasets adaptations demonstrate that the proposed large margin domain adaptation algorithm achieves superior classification accuracy comparing with the state of the art methods. Quantitative evaluation of the proposed algorithm shows that semi-supervised adaptation utilizing one sample per class of target domain set reduces the error rates by 64.72% comparing with a corresponding SVM classifier.     

Hyperspectral data analysis and supervised feature reduction viaprojection pursuit

As the number of spectral bands of high-spectral resolution data increases, the ability to detect more detailed classes should also increase, and the classification accuracy should increase as well. Often the number of labelled samples used for supervised classification techniques is limited, thus limiting the precision with which class characteristics can be estimated. As the number of spectral bands becomes large, the limitation on performance imposed by the limited number of training samples can become severe. A number of techniques for case-specific feature extraction have been developed to reduce dimensionality without loss of class separability. Most of these techniques require the estimation of statistics at full dimensionality in order to extract relevant features for classification. If the number of training samples is not adequately large, the estimation of parameters in high-dimensional data will not be accurate enough. As a result, the estimated features may not be as effective as they could be. This suggests the need for reducing the dimensionality via a preprocessing method that takes into consideration high-dimensional feature-space properties. Such reduction should enable the estimation of feature-extraction parameters to be more accurate. Using a technique referred to as projection pursuit (PP), such an algorithm has been developed. This technique is able to bypass many of the problems of the limitation of small numbers of training samples by making the computations in a lower-dimensional space, and optimizing a function called the projection index. A current limitation of this method is that, as the number of dimensions increases, it is likely that a local maximum of the projection index will be found that does not enable one to fully exploit hyperspectral-data capabilities     

An Equivalent Form of S-L Projection Learning

Learning a mapping between an input space and an output.space from training data canbe viewed as a problem of function approximation,which means that different criteria result in learningapproaches of different abilities.Among them,projection criterion proposed by one of the authors aimsdirectly at optimal generalization.The projection idea leads to three specific learning approaches,projection learning,partial projection learning,and averaged projection learning,and a framework of afamily of projection learning called S-L projection learning is established to discuss infinite kinds oflearning.S-L projection learning has a different form from the three methods,and it is not easy to analyzetheir relationship.This paper focuses on an equivalent form of S-L projection learning,which shows thatit is a transformed version of partial projection learning.     

基于pinball损失的结构模糊孪生支持向量机

孪生支持向量机通过求解较小的二次规划问题,提高了分类器的性能,然而,该方法主要利用了类间可分的特性,并使用hinge损失函数构建相应的模型,它们并未充分考虑不同类中数据的结构信息以及不同样本对分类的影响,导致该方法对噪声具有较强的敏感性以及重取样的不稳定性.为了进一步提高孪生支持向量机的性能,基于pinball损失函数,将数据集中不同类的结构信息以及不同样本的作用引入到孪生支持向量机中,获得了基于pinball损失的结构模糊孪生支持向量机模型,从理论上导出了基于pinball损失的结构模糊孪生支持向量机算法pin-sftsvm,通过选取人工生成数据集与UCI标准数据集,对pin-sftsvm算法进行了实验,并与tbsvm、s-tsvm和pin-tsvm算法进行了性能比较,表明了提出算法的有效性.     

基于Pinball损失函数支持向量机的极化SAR图像鲁棒分类

考虑到极化合成孔径雷达(PolSAR)图像标注信息量低以及相干斑噪声难以消除的问题,该文从鲁棒统计学习的角度提出了一种基于Pin-SVM的极化SAR图像鲁棒分类方法,根据极化SAR图像的散射特性和地物的纹理特性,通过求解两类样本之间的最大分位数距离来确定分类超平面,在无需迭代的前提下得到更加鲁棒的分类结果。相比传统的基于最大间隔的极化SAR图像分类算法,该文所提算法一方面在对极化SAR图像提取到的特征中包含的噪声具有更好的鲁棒性,另一方面对于训练样本的抽样范围不敏感,即重采样具有更好的鲁棒性。利用EMISAR的Foulum地区极化SAR数据进行了算法验证,多种情况的对比实验的结果验证了该算法的有效性。      

多方计算任务的量子通信复杂度

假设多个用户分别根据各自持有的函数对共享数据进行计算,用户之间采用互相通信的方式完成一个共同的目标任务。本文基于一个通用的判别函数模型,给出对于上述任务,采用经典最优算法下的经典通信复杂度。然后,以Grover算法为基础,文中构造了适用于前述任务的量子分布式算法,并给出相应的量子通信复杂度。研究表明,量子算法的性能取决于函数定义域与用户数的无穷大阶的差距。量子通信复杂度较之经典情形最多将会有二次级别的降低。      

一种新的量子神经网络训练算法

量子神经网络是一种借鉴量子理论中的态叠加思想而设计的单隐层前馈神经网络,其主要用于数据分类。由于采用多层激励函数神经元,并且在量子间隔训练中采用了新的目标函数,即同类输入数据的隐层节点输出方差最小,从而使量子神经网络具备了发掘不同类别数据间模糊性的能力。但由于训练时对量子神经网络权值和量子间隔使用了不同的目标函数,使迭代过程中两者不可避免的会出现相互冲突,从而导致训练迭代次数的增加和网络性能的下降。本文借鉴约束优化理论,在两个目标函数的梯度下降求解中引入了惩罚函数,提出了一种新的量子神经网络训练算法,消除了两个目标函数间的冲突。实验结果表明,本文提出的训练算法可以显著提升训练的速度和网络的性能。      

Robust L1-norm two-dimensional collaborative representation-based projection for dimensionality reduction

Collaborative representation-based projection (CRP) is a well-known dimensionality reduction technique, which has been proved to have better performance than sparse representation-based projection (SRP) in the fields of recognition and computer vision. However, classical CRP is sensitive to noises and outliers since its objective function is based on L2-norm, and it will suffer from the curse of dimensionality as it is used for images processing. In this paper, a novel CRP model, named L1-norm two-dimensional collaborative representation-based projection (L1-2DCRP) and an efficient iterative algorithm to solve it are proposed. Different from conventional CRP, the optimal problem in our proposed model is a L1-norm-based maximization and the vector data is extended to matrix date. The proposed algorithm is theoretically proved to be monotonously convergent, and more robust to noises and outliers since L1-norm is used. Experimental results on CMU Multi-PIE, COIL20, FERET and ORL face databases validate the effectiveness of L1-2DCRP compared with several state-of-the-art approaches.     

基于支持向量机的Web文本分类方法

Web文本分类技术是数据挖掘中一个研究热点领域，而支持向量机又是一种高效的分类识别方法，在解决高维模式识别问题中表现出许多特有的优势。文章通过分析Web文本的特点，研究了向量空间模型（VSM）的分类方法和核函数的选取，在此基础上结合决策树方法提出了一种基于决策树支持向量机的Web文本分类模型。并给出具体的算法。通过实验测试表明，该方法训练数据规模大大减少，训练效率较高，同时具有较好的精确率（90．11％）和召回率（89．38％）。     

基于粒子群优化的直觉模糊核匹配追踪算法

针对现有直觉模糊核匹配追踪算法采用贪婪算法搜索最优基函数而导致学习时间过长的问题,汲取了粒子群优化算法全局搜索能力强、收敛速度快的优势对最优基函数的搜索过程进行优化,提出了一种基于粒子群优化的直觉模糊核匹配追踪算法,并将该算法应用于时效性要求更高的空天目标识别领域.实验结果表明,与传统方法相比,本文方法在识别率相当的情况下有效缩短一次匹配追踪时间,计算效率明显提高,且所得模型具有稀疏性好,泛化能力高等优点,特别适用于兼顾识别率和实时性的应用领域.     

基于新型三元卷积神经网络的行人再辨识算法

基于三元卷积神经网络的行人再辨识算法多数采用欧式距离度量行人之间的相似度,并配合铰链(hinge)损失函数进行卷积神经网络的训练。然而,这种作法存在两个不足：欧式距离作为行人相似度,鉴别力不够强;铰链损失函数的间隔(Margin)参数设定依赖于人工预先设定且在训练过程中无法自适应调整。为此,针对上述两个不足进行改进,该文提出一种基于新型三元卷积神经网络的行人再辨识算法,以提高行人再辨识的准确率。首先,提出一种归一化混合度量函数取代传统的度量方法进行行人相似度计算,提高了行人相似度度量的鉴别力;其次,提出采用Log-logistic函数代替铰链函数,无需人工设定间隔参数,改进了特征与度量函数的联合优化效果。实验结果表明,所提出的算法在Auto Detected CUHK03 和VIPeR两个数据库上的准确率均获得显著的提升,验证了所提出算法的优越性。     

基于多样本的多核Fisher判别分析研究

多核Fisher判别分析法是一种有效的非线性判别分析法,对其涉及的参数利用遗传算法进行确定是一个有效的途径。针对训练样本较多时遗传算法搜索时间较长的问题,提出一种基于多样本的多核Fisher算法。其做法是将犬的训练集拆分成若干个小样本集,依次求得投影映射,并利用“投票策略”来判别待测样本。在人脸识别上的实验表明,基于多样本的多核Fisher算法可以在不降低分类正确率的前提下,提高算法的运算速度。     

基于改进的蛙跳算法与支持向量机的实用语音情感识别

支持向量机是一种基于统计学习理论的新型机器学习算法,在高维特征空间中寻找最优分类超平面,具有很好的分类精度和泛化性能。支持向量机的训练需要求解一个带约束的二次规划问题,针对数据规模很大情况下经典训练方法变得很困难的缺点,提出一种基于改进混合蛙跳算法的支持向量机分类器参数优化方法,既提高了混合蛙跳算法的收敛速度和精度,又能借助混合蛙跳算法的全局随机搜索能力,为支持向量机参数的优化选择提供一条有效途径。本文提取情感语句的韵律特征、音质特征和混沌特征参数,提出一种基于改进蛙跳算法的数据融合方法,并利用基于改进蛙跳算法的支持向量机进行实用语音情感的识别研究。在实验中比较了PCA方法、BP神经网络和数据融合方法用于语音情感识别的识别率,研究结果表明本文所提的各项改进机制能有效提升情感识别率,为实用语音情感的识别提供一种新方法和新思路。      

变参数QPSO算法优化神经网络的短期电力负荷预测

为了提高电力负荷预测精度,提出了一种变参数量子粒子群(VPQPSO)算法优化RBF神经网络的短期负荷预测模型(VPQPSO-RBFNN)。首先利用电力负荷的混沌性,对短期负荷时间序列进行相空间重构;然后采用变参数QPSO算法优化RBF神经网络参数对重构后的短期负荷时间序列进行学习,建立短期电力负荷最优预测模型;最后采用对某地区短期电力负荷进行预测。VPQPSO-RBFNN可以准确描述复杂多变的电力负荷变化趋势,提高了电力负荷的预测精度,仿真结果验证了VPQPSO-RBFNN可以用于电力系统负荷预测。     

面向软件定义网络的服务功能链优化部署算法研究

针对网络功能虚拟化(NFV)环境下,现有服务功能链部署方法无法在优化映射代价的同时保证服务路径时延的问题,该文提出一种基于IQGA-Viterbi学习算法的服务功能链优化部署方法。在隐马尔可夫模型参数训练过程中,针对传统Baum-Welch算法训练网络参数容易陷入局部最优的缺陷,改进量子遗传算法对模型参数进行训练优化,在每一迭代周期内通过等比例复制适应度最佳种群的方式,保持可行解多样性和扩大空间搜索范围,进一步提高模型参数的精确度。在隐马尔科夫链求解过程中,针对隐含序列无法直接观测这一难点,利用Viterbi算法能精确求解隐含序列的优势,解决有向图网络中服务路径的优化选择问题。仿真实验结果表明,与其它部署算法相比,所提IQGA-Viterbi学习算法能有效降低网络时延和映射代价的同时,提高了网络服务的请求接受率。

     

基于样本扩张和最大散度差融合的单样本人脸识别算法

为解决只有一个训练样本时最大散度差(MSD) 鉴别分析在人脸识别中的识别性能会降低这一问题,提 出一种基于样本扩张和MSD融合的单样本人脸识别算法。首先,根据人脸的对称相似理论 ,人脸样本的相关变化信息可 以从它的对称脸上提取,并且平均脸也具有要识别测试人脸的某些可能变化,提出组合原始 训练样本及它的虚 拟平均脸和虚拟对称脸作为新的训练样本集;然后,在新的训练样本集上应用类内中间值MS D鉴别分析算法得到最 优投影矩阵,从而可以基于这个最优投影矩阵计算训练样本和待测试样本的特征;最后利用 模糊决策方法进行分类。在ORL和FERET人脸数据库上的大量实验结果表明,本文算法可以提 高识别率,具有一定的鲁棒性。