新型异步电机无速度传感器低速控制算法

基于数学模型的速度估算是异步电机无速度传感器控制的核心。目前,围绕速度估算这个问题,在电机控制领域已出现模型参考自适应、自适应观测器以及扩展卡尔曼滤波等多种方法。无论采用哪种速度估算技术,速度估算系统的动静态性能、低速性能、对参数的敏感性、算法的复杂程度及实现难度都是工程师必须考虑的几个重要性能指标。文中针对低速条件下的异步电机速度估算及控制提出了一种基于无功功率闭环的异步电机无速度传感器矢量控制算法,并探讨了该方法的一些性能指标问题。     

MapReduce optimization algorithm based on machine learning in heterogeneous cloud environment

We present an approach to optimize the MapReduce architecture, which could make heterogeneous cloud environment more stable and efficient. Fundamentally different from previous methods, our approach introduces the machine learning technique into MapReduce framework, and dynamically improve MapReduce algorithm according to the statistics result of machine learning. There are three main aspects： learning machine performance, reduce task assignment algorithm based on learning result, and speculative execution optimization mechanism. Furthermore, there are two important features in our approach. First, the MapReduce framework can obtain nodes＇ performance values in the cluster through machine learning module. And machine learning module will daily calibrate nodes＇ performance values to make an accurate assessment of cluster performance. Second, with the optimization of tasks assignment algorithm, we can maximize the performance of heterogeneous clusters. According to our evaluation result, the cluster performance could have 19% improvement in current heterogeneous cloud environment, and the stability of cluster has greatly enhanced.     

改进蝙蝠算法优化极限学习机的图像分类

针对分类器中的极限学习机参数优化问题,本文提出一种改进蝙蝠算法优化极限学习机的图像分类模型。首先将极限学习机参数看作蝙蝠位置,然后采用改进蝙蝠算法进行求解。采用病毒群体感染主群体,主群体在历代个体间纵向传递信息,病毒群体通过感染操作在同代个体间横向传递信息,增强了算法跳出局部极小值的能力。最后根据最优参数建立图像分类模型,并对模型的性能进行仿真测试。仿真结果表明,相对于对比模型,本文模型不仅提高了图像分类正确率,而且加快了分类速度,是一种有效的图像分类模型。     

基于机器学习的图像分割算法研究

图像分割是计算机视觉研究中重要的一部分,其主要目的是在图像中将兴趣域目标与背景分割,关系到后续的目标识别、图像理解等操作的准确性.经过几十年的发展,许多优秀的图像分割的方法被提出.机器学习是当今时代的研究热点,基于深度卷积神经网络等机器学习的图像分割研究进展迅速.总结介绍了应用于图像分割的几种典型机器学习方法,分析比较了相关的分割原理步骤、优缺点和发展现状.最后分析了基于机器学习的图像分割算法的发展方向.     

Regularized manifold information extreme learning machine

By exploiting the thought of manifold learning and its theoretical method, a regularized manifold information ex-treme learning machine algorithm aimed to depict and fully utilize manifold information was proposed. The proposed algo-rithm exploited the geometry and discrimination manifold information of data to perform network of ELM. The proposed algorithm could overcome the problem of the overlap of information. Singular problems of inter-class and within-class were solved effectively by using maximum margin criterion. The problem of inadequate learning with limited samples was solved. In order to demonstrate the effectiveness, comparative experiments with ELM and the related update algorithms RAFELM, GELM were conducted using the commonly used image data. Experimental results show that the proposed algorithm can significantly improve the generalization performance of ELM and outperforms the related update algorithms.     

基于有序编码的核极限学习顺序回归模型

顺序回归是机器学习领域中介于分类和回归之间的有监督问题。在实际中,许多带有序关系标签的问题都可以被建模成顺序回归问题,因此顺序回归受到众多学者的关注。基于极限学习机(ELM)的算法能有效避免因迭代过程陷入的局部最优解,减少训练时间,但基于极限学习机的算法在顺序回归问题上的研究较少。该文将核极限学习机与纠错输出编码相结合,提出了一种基于有序编码的核极限学习顺序回归模型。该模型有效解决了如何在顺序回归中取得良好的特征映射以及如何避免传统极限学习机中隐层节点个数依赖于人工设置的问题。为验证提出模型的有效性,该文在多个顺序回归数据集上进行了测试,测试结果表明,相比于传统ELM模型,该文提出的模型在准确率上平均提升了10.8%,在数据集上预测表现最优,而且获得了最短的训练时间,从而验证了模型的有效性。     

A learning‐based approach for virtual machine placement in cloud data centers

In recent years, the increasing use of cloud services has led to the growth and importance of developing cloud data centers. One of the challenging issues in the cloud environments is high energy consumption in data centers, which has been ignored in the corporate competition for developing cloud data centers. The most important problems of using large cloud data centers are high energy costs and greenhouse gas emission. So, researchers are now struggling to find an effective approach to decreasing energy consumption in cloud data centers. One of the preferred techniques for reducing energy consumption is the virtual machines (VMs) placement. In this paper, we present a VM allocation algorithm to reduce energy consumption and Service Level Agreement Violation (SLAV). The proposed algorithm is based on best‐fit decreasing algorithm, which uses learning automata theory, correlation coefficient, and ensemble prediction algorithm to make better decisions in VM allocation. The experimental results indicated improvement regarding energy consumption and SLAV, compared with well‐familiar baseline VM allocation algorithms.     

一种稳定的基于极端学习机的纹理分类方法

针对传统基于极端学习机(ELM)的纹理分类方 法容易出现输出不稳定的缺点,本文将线性和非线 性模型进行融合,改进了传统动力学模型。利用ELM能够快速学习的特点,将其作为基分类 器,同时利用 线性吸引子和局部吸引子的迭代,实现多个ELM分类器的最佳融合,以提升学习模型的泛 化能力和稳定 性。利用动力模型实现多分类器的融合有助于寻求多个基分类器之间的一致性,摒弃了基分 类器中判别错 误的样本输出。通过对CUReT纹理数据库的分类结果,与传统纹理分类方法相 比,本文算法的稳定性 和分类准确率都得到了一定程度的提升,取得了理想的分类效果。     

A federated semi-supervised learning approach for network traffic classification

The classification of network traffic, which involves classifying and identifying the type of network traffic, is the most fundamental step to network service improvement and modern network management. Classic machine learning and deep learning methods have widely adopted in the field of network traffic classification. However, there are two major challenges in practice. One is the user privacy concern in cross-domain traffic data sharing for the purpose of training a global classification model, and the other is the difficulty to obtain large amount of labeled data for training. In this paper, we propose a novel approach using federated semi-supervised learning for network traffic classification, in which the federated server and clients from different domains work together to train a global classification model. Among them, unlabeled data are used on the client side, and labeled data are used on the server side. The experimental results derived from a public dataset show that the accuracy of the proposed approach can reach 97.81%, and the accuracy gap between the federated learning approach and the centralized training method is minimal.     

A supervised learning approach to automate the acquisition of knowledge in surveillance systems

This paper presents a machine learning-based method to build knowledge bases used to carry out surveillance tasks in environments monitored with video cameras. The method generates three sets of rules for each camera that allow to detect objects’ anomalous behaviours depending on three parameters: object class, object position, and object speed. To deal with uncertainty and vagueness inherent in video surveillance we make use of fuzzy logic. Thanks to this approach we are able to generate a set of rules highly interpretable by security experts. Besides, the simplicity of the surveillance system offers high efficiency and short response time. The process of building the knowledge base and how to apply the generated sets of fuzzy rules is described in depth for a real environment.     

机器学习技术在医疗数据挖掘中的应用

随着医疗事业的不断发展,我国医疗技术也得到了有效的进步,医疗技术在不断完善。这其中,医疗测量仪器技术得到了显著的发展,能够更好地支持医疗信息记录工作。本文通过对医疗数据挖掘的相关问题进行研究,指出了医疗数据挖掘的发展方向。     

Automatic and objective gradation of 114 183 terrorist attacks using a machine learning approach

Catastrophic events cause casualties, damage property, and lead to huge social impacts. To build common standards and facilitate international communications regarding disasters, the relevant authorities in social management rank them in subjectively imposed terms such as direct economic losses and loss of life. Terrorist attacks involving uncertain human factors, which are roughly graded based on the rule of property damage, are even more difficult to interpret and assess. In this paper, we collected 114 183 open-source records of terrorist attacks and used a machine learning method to grade them synthetically in an automatic and objective way. No subjective claims or personal preferences were involved in the grading, and each derived common factor contains the comprehensive and rich information of many variables. Our work presents a new automatic ranking approach and is suitable for a broad range of gradation problems. Furthermore, we can use this model to grade all such attacks globally and visualize them to provide new insights.     

带视频输出学习机的设计

针对目前学习机屏幕小,功能内容单一,智能化不强,单声道发音,个人单用等特点,提出一种带视频输出的学习机,使用了全新的设计理念,图文并茂,声音丰富,智能程度高,将学习机里的内容通过外接电视机或投影设备显示出来,通过大屏幕显示可让多人互动学习。省去了自带显示屏的成本和避免了屏幕显示太小的问题。     

面向机器学习的隐私保护关键技术研究综述

随着信息通信技术的发展,机器学习已经成为多个研究领域与垂直行业必不可少的技术工具。然而,机器学习所需数据中往往包含了大量的个人信息,使其隐私保护面临风险与挑战,受到了越来越多的关注。对现有机器学习下隐私保护法规政策与标准化现状进行梳理,对适用于机器学习的隐私保护技术进行详细介绍与分析。隐私保护算法通常会对数据质量、通信开支与模型表现等造成影响,因此对于隐私保护算法的评估应当进行多维度的综合评估。总结了适用于机器学习应用的隐私保护性能评估指标,并指出隐私保护需要考虑对数据质量、通信开支以及模型准确率等之间的影响。     

Hybrid beamforming design based on unsupervised machine learning for millimeter wave systems

This article proposes a hybrid beamforming design with reduced channel state information (CSI) feedback. We use a beam sweeping procedure to provide channel measurements and feed a CSI report scheme. Thereby, the base station (BS) can perform an adequate estimation of the channel characteristics with reduced signaling overhead. Consequently, we need short pilot sequences and very few precoding matrix indicators (PMIs) to properly describe channel behavior. Moreover, we also evaluate different user selection strategies based on unsupervised machine learning framework that exploits the channel information provided by the proposed beam sweeping scheme. Our performance evaluation indicates that the user selection based on fuzzy c‐means is able to efficiently explore the reduced CSI. The proposed hybrid beamforming scheme reduces the multi‐user interference and achieves significant gains in total data rate as channel conditions and interference environment becomes more challenging.     

面向机器学习模型的基于PCA的成员推理攻击

针对目前黑盒成员推理攻击存在的访问受限失效问题,提出基于主成分分析(PCA)的成员推理攻击.首先,针对黑盒成员推理攻击存在的访问受限问题,提出一种快速决策成员推理攻击fast-attack.在基于距离符号梯度获取扰动样本的基础上将扰动难度映射到距离范畴来进行成员推理.其次,针对快速决策成员推理攻击存在的低迁移率问题,提...     

基于支持向量机的粗糙海面风速及海表盐度反演研究

将支持向量机(Support Vector Machine, SVM)回归技术应用到海况参数(如海表盐度、海面风速等)反演研究.利用双尺度模型(Two-Scale Model, TSM)作为前向电磁算法, 数值模拟不同雷达参数下风驱粗糙海面微波后向散射系数, 经过敏感性分析, 选取L波段(1.4 GHz)、C波段(6.8 GHz)及其合适的入射角作为雷达参数, 并设计多种反演方案, 分别以单频率双极化双角度、双频率双极化双角度及双极化后向散射系数的比值作为SVM的训练样本数据信息, 经过适当的训练, 利用SVM回归技术对海洋表面风速和盐度进行了反演研究.研究结果表明, 针对于海面风速的反演, C波段的反演精度最高, 针对于海表盐度的反演, L波段同极化散射系数比值作为SVM输入的反演精度较高.最后, 检验了SVM反演方法的抗噪声性能, 表明文中提出的SVM方法能较好地应用于实际海况参数反演问题.     

新的基于机器学习的入侵检测方法

提出了一种基于机器学习的用户行为异常检测方法,主要用于UNIX平台上以shell命令为审计数据的入侵检测系统。该方法在LaneT等人提出的检测方法的基础上,改进了对用户行为模式和行为轮廓的表示方式,在检测中以行为模式所对应的命令序列为单位进行相似度赋值;在对相似度流进行平滑时,引入了“可变窗长度”的概念,并联合采用多个判决门限对被监测用户的行为进行判决。实验表明,该方法在检测准确度和实时性上均优于LaneT等人提出的方法。     

云计算环境下可信虚拟机管理模型

为了解决云计算环境下虚拟机管理存在的管理域特权过于集中和用户策略易被恶意篡改等问题,提出了一种可信虚拟机管理模型。模型首先对虚拟机管理域进行了细粒度的划分,赋予管理员和用户不同的管理特权,防止管理员随意访问用户的数据;利用可信计算技术建立可信通道分发用户策略,防止管理员恶意篡改用户策略。安全性分析与实验测试表明,该模型可以有效保护用户数据和用户策略的安全性。     

基于新型机器学习方法的蛋白质功能预测与分析

介绍了蛋白质功能预测与分析的现状和主要研究内容,并说明了如何利用新型机器学习方法来进行这方面的研究工作,并对其以后的发展提出了展望.