非监督子空间学习中关联度量多维尺度分析

多维尺度分析已经在维度约减和数据挖掘领域得到了广泛应用。MDS的主要缺点是其定义在训练数据上,对于新的测试样本无法直接获得映射结果。另外,MDS基于欧氏距离度量,不适合获取相似数据中的非线性流形结构。将MDS扩展到关联度量空间,称为关联度量多维尺度分析(CMDS)。与传统MDS在训练数据中完成映射,进而缩小空间范围相比,CMDS 能够直接获得测试样本映射结果。此外,CMDS基于关联度量,能够有效学习相似数据中的非线性流形结构。理论分析表明,CMDS可以利用核方法扩展到新特征空间,解决非线性问题。实验结果表明,CMDS及其核形式KG-CMDS性能优于常用传统降维方法。     

基于多维定标的无线传感器网络三维定位算法*

在森林防火、目标追踪、灾难预警、环境监测等应用中,需要通过定位算法对无线传感器节点进行三维定位。提出一种基于多维定标的无线传感器网络三维定位算法,结合RSS经验衰减模型和最短路径建立相异性矩阵,采用轻量级矩阵分解算法降低相异性矩阵分解的计算复杂性,并利用网络中存在的周期性消息将初始定位信息回送,在后台使用迭代优化算法对初始定位结果求精。仿真实验表明,在测距误差一定的情况下,该算法能够提高节点三维坐标的初始计算精度,经过集中式的优化求精后与MDS-MAP算法相比,能够明显地提高节点三维定位的精度。     

基于粒子群优化的多维标度节点定位算法

针对无线传感网络节点定位易受外部环境影响的问题,提出一种基于粒子群寻优的多位标度定位算法。利用基于动态路径损耗指数的接受信号强度测距,建立距离矩阵,使用多维标度方法构建节点的相对坐标,通过四参数坐标转换模型得到绝对坐标,再用绝对坐标与实际坐标的差异度作为粒子群寻优的适应度函数,通过分群搜索,优化节点位置估计。仿真结果表明：改进算法对实际环境影响具有较好的鲁棒性,节点定位精度有了明显的提高,能够满足定位系统需求。     

Comparative aspects of software complexity metrics and program modules — a multidimensional scaling approach

A multivariate statistical procedure called multidimensional scaling is used to study the relationship of various software complexity metrics and program modules. The program modules that make up a software system are analysed and their effects towards the overall characteristics of a software are viewed. This multidimensional scaling technique is applied to a sample data set. The scaling procedure clustered the similar and dissimilar software complexity metrics. Program modules with low complexity and few errors clustered together, while modules which were complex were isolated. This technique shows promise in the identification of complex modules that potentially contain disproportionate errors prior to the testing phase. The ability of the scaling techniques to cluster similar and dissimilar characteristics is explained and graphically presented.     

Visual exploration of data by using multidimensional scaling on multicore CPU,GPU, and MPI cluster

Visual and interactive data exploration requires fast and reliable tools for embedding of an original data space in 3(2)‐dimensional Euclidean space. Multidimensional scaling (MDS) is a good candidate. However, owing to at least O(M²) memory and time complexity, MDS is computationally demanding for interactive visualization of data sets consisting of order of 10⁴ objects on computer systems, ranging from PC with multicore CPU processor, graphics processing unit (GPU) board to midrange MPI clusters. To explore interactively data sets of that size, we have developed novel efficient parallel algorithms for MDS mapping based on virtual particle dynamics. We demonstrate that the performance of our MDS algorithms implemented in compute unified device architecture environment on a PC equipped with a modern GPU board (Tesla M2090, GeForce GTX 480) is considerably faster than its MPI/OpenMP parallel implementation on the modern midrange professional cluster (10 nodes, each equipped with 2x Intel Xeon X5670 CPUs). We also show that the hybridized two‐level MPI/CUDA implementation, run on a cluster of GPU nodes, can additionally provide a linear speedup. Copyright 2013 John Wiley & Sons, Ltd.     

自动车辆识别与装车系统研究

针对煤矿汽车装车过程中司机频繁下车交票、进出控制室造成的车辆通行慢、效率低、人为作弊等问题,设计了适合汽车运输、计量、装车的自动车辆识别和装车系统。该系统实现了对运输车辆的监测管理、自动识别、连续装车、自动记录等功能,缩短了装车辅助时间,提高了煤矿运输效率和安全度。     

Potential applications of three-way multidimensional scaling and related techniques to integrate knowledge from multiple experts

Theknowledge transfer problem in artificial intelligence consists of finding effective ways to elicit information from a human expert and represent it in a form suitable for use by an expert system. One approach to formalizing and guiding this knowledge transfer process for certain types of expert systems is to use psychometric scaling methods to analyze data on how the human expert compares or groups solutions. For example, Butler and Corter [1] obtained judgments of thesubstitutability of solutions from an expert, then analyzed the resulting data via techniques for fitting trees and extended trees [2]. The expert's interpretation of certain aspects of the solutions were directly encoded as production rules, allowing rapid prototyping. In this paper we consider the problem of combining information from multiple experts. We propose the use of three-way or individual differences multidimensional scaling, tree-fitting, and unfolding models to analyze two types of data obtainable from the multiple experts: judgments of the substitutability of pairs of solutions, and judgments of the appropriateness of specific solutions to specific problems. An application is described in which substitutability data were obtained from three experts and analyzed using the SINDSCAL program [3] for three-way multidimensional scaling [4].     

The effects of gender,flow and video game experience on combat identification training

Abstract

The present study examined the effects of gender, video game experience (VGE), and flow state on multiple indices of combat identification (CID) performance. Individuals were trained on six combat vehicles in a simulation, presented through either a stereoscopic or non-stereoscopic display. Participants then reported flow state, VGE and were tested on their ability to discriminate friend vs. foe and identify both pictures and videos of the trained vehicles. The effect of stereoscopy was not significant. There was an effect of gender across three dependent measures. For the two picture-based measures, the effect of gender was mediated by VGE. Additionally, the effect of gender was moderated by flow state on the identification measures. Overall, the study suggests that gender differences may be overcome by VGE and by achieving flow state. Selection based on these individual differences may be useful for future military simulation.

Practitioner Summary: This work investigates the effect of gender, VGE and flow state on CID performance. For three measures of performance, there was a main effect of gender. Gender was mediated by previous VGE on two measures, and gender was moderated by flow state on two measures.     

Heavy vehicle suspension parameters identification and estimation of vertical forces: experimental results

The aim of the present work is to estimate the vertical forces of heavy vehicle and identify the unknown dynamic parameters using sliding mode observer approach. This observation needs a good knowledge of dynamic parameters such as damping coefficient, spring stiffness, etc. In this paper, suspension stiffness and unsprung masses have been identified. Experimental results carried out on an instrumented tractor have been presented in order to show the quality of the state observation, parameters identification and force estimation. These estimation results are then compared to the measured one coming from the sensors installed in the tractor. Many scenarios have been tested. In this paper, the results coming from zigzag test have been shown and commented.     

智能导师系统在军事训练中应用现状及发展

根据国内当前军事训练所面临的实际困境及智能导师系统(ITS)在教学训练应用中所具有的功能和经济价值,对ITS在军事训练中的应用进行了初步研究.介绍了ITS的发展历程、ITS的相关概念及其技术特点.以ITS在国外应用及发展为主线,重点论述了欧美国家军事训练中ITS应用的现状和相关的典型系统.提出在军事训练领域中我们应用ITS技术的一些建议.     

联合白噪声统计估计和卡尔曼滤波的MIMO信道辨识

提出了一种新的时间选择性衰落环境下MIMO信道辨识算法。为了提高信息传输效率,训练序列被直接叠加于信息序列之上。算法将信息符号输出、接收端AWGN和由于采用零中频接收技术而产生的直流偏移当做虚拟的观测噪声,其均值和自协方差均未知。通过联合的递推白噪声统计估计器和卡尔曼滤波器对时变信道进行跟踪,推导了一种计算简单的次优无偏时变白噪声统计估计器。以简单有效的方法抑制直流偏移对辨识精度的影响。仿真结果表明了算法具有良好的性能。     

Scaled conjugate gradient and Bayesian training of neural networks for fault identification in cylinders

This paper compares Bayesian training of neural networks using hybrid Monte Carlo to scaled conjugate gradient method for fault identification in cylinders using vibration data. From the measured data pseudo-modal energies and modal properties are calculated and the coordinate pseudo-modal energy assurance criterion (COMEAC) and the coordinate modal assurance criterion (COMAC) are computed respectively. The pseudo-modal energies, modal properties, COMEAC and COMAC are used to train four neural networks. On average, the pseudo-modal-energy-networks are more accurate than the modal-property-networks. The weighted averages of the pseudo-modal-energy- and modal-property-networks form a committee of networks. The committee method gives lower mean squared errors and better classification of faults than the individual methods. The Bayesian training is found to be more accurate and computationally expensive than the scaled conjugate gradient method and to give confidence levels.     

Behaviour modelling,instruction and exploration training approaches in group and individual contexts

Behaviour modelling has been associated with higher learning outcomes compared to other training approaches. These cumulative research findings create imperative to examine underlying causal mechanisms or contingency factors that may promote behaviour modelling's advantages even further. We propose group-based learning as one contingency factor because there exists greater opportunity for observation, imitation and feedback. We use a two-by-three experimental laboratory design involving 84 subjects to test for an interaction effect between training context (group-based, individual-based) and training approach (behaviour modelling, exploration, instruction) on several learning outcomes. We use hierarchical regression to show that, while the interaction effect was not significant, the training approach main effect was significant. Consistent with expectations, behaviour modelling outperformed exploration on task performance. There were no significant differences between behaviour modelling and instruction. We conclude that behaviour modelling is associated with higher task performance levels on a complex word-processing task when compared to exploration.     

正规化模糊神经网络及在手写体汉字识别中的应用

为改善手写体汉字识别的性能,提出了一种基于正规化模糊神经网络的识别方法。针对网络结构的优化问题给出了网络模型的规则层节点的选取方法和相应的反传播学习规则。该算法能够充分利用专家制订的“if-then”规则,完善网络的推理结构,提高网络的识别能力,减少噪声因素的影响。实验表明此方法对手写体汉字识别问题具有良好的适应性和实用性。该方法指出了一条进一步提高手写体汉字系统性能的新途径。     

基于可见光视觉图像的路面裂缝识别深度学习方法述评

基于可见光视觉图像的表面裂缝识别为非接触式,不受被测对象材质限制,可在线自动检测,具有速度快、成本低和精度高等优势。首先较为全面地搜集了典型的路面裂缝公开数据集,整理归纳了样本特征及其随机可变影响因素,并比较了传统手工设计特征工程、机器学习和深度学习3种主要裂缝识别方法的优缺点。然后,从网络架构、性能和效果方面着重评述了自搭架构、迁移学习和编码-解码器等易于训练和部署的深度学习算法新进展,通过算法优化和算力提升可显著提高识别的效果和性能,测试结果表明能够在低算力平台上实现裂缝补丁级快速检测和像素级实时检测。     

The impact of synchronous inter-networked teacher training in Information and Communication Technology integration

This paper summarises the outcomes of research that recorded and interpreted change in pre-service teachers’ pedagogical beliefs when immersed in unique synchronous networked collaboration. The pre-service teachers were involved in the design and implementation of synchronous networked tasks, supported by the theoretical and pragmatic application of ‘informed’ Information and Communications Technology (ICT) integration. The pre-service teachers’ use of digital technology graduated from a didactic, broadcast delivery of information to more constructivist strategies that involved interaction, collaboration and ‘good’ learning. Moreover, the pre-service teachers gained competencies during a 12-week programme that led to the development of a suggested framework for ICT task development able to effectively promote and support improvements in learning.     

The Xanadu project: training faculty in the use of information and communication technology for university teaching

Abstract Technology‐enhanced learning (TEL) methodologies are becoming an important part of University teaching but faculty members have tended to shy away from using them. So, how can they be enticed to use them effectively? What approaches can be used? The purpose of the Xanadu project was to analyse the problems involved and to propose a model for training, based on experimentation at the University of Turin. Besides providing a model, this article will deal with the follow‐up, particularly regarding the faculty members' initial approach to using information and communication technology (ICT), as well as examine the motives and conditions influencing their choices. In this sense, the project has enabled us to come to a better understanding of the typical misconceptions leading teachers to favour methodological approaches based on the distributive use of ICT (which are considered less demanding to manage), rather than networked collaborative interaction, which actually involve students more. In this regard, Xanadu has shown how teachers' awareness may be developed towards adopting a wide range of TEL approaches through both gradual training (project‐oriented with a basic and an advanced course) and with the help of a graduate assistant capable of following e‐content development and online collaborative activities. The effectiveness of the method may be confirmed by the large number of faculty members continuing to use ICT to support their teaching despite having no specific university TEL projects.     

Motivational and cognitive benefits of training in immersive virtual reality based on multiple assessments

The main objective of this study was to examine the effectiveness of immersive virtual reality (VR) as a medium for delivering laboratory safety training. We specifically compare an immersive VR simulation, a desktop VR simulation, and a conventional safety manual. The sample included 105 first year undergraduate engineering students (56 females).  We include five types of learning outcomes including post‐test enjoyment ratings; pre‐ to post‐test changes in intrinsic motivation and self‐efficacy; a post‐test multiple choice retention test; and two behavioral transfer tests. Results indicated that the groups did not differ on the immediate retention test, suggesting that all three media were equivalent in conveying the basic knowledge. However, significant differences were observed favoring the immersive VR group compared to the text group on the two transfer tests involving the solving problems in a physical lab setting (d = 0.54, d = 0.57), as well as enjoyment (d = 1.44) and increases in intrinsic motivation (d = 0.69) and self‐efficacy (d = 0.60). The desktop VR group scored significantly higher than the text group on one transfer test (d = 0.63) but not the other (d= 0.11), as well as enjoyment (d =1.11) and intrinsic motivation (d =0.83).     

Simultaneous state estimation and parameter identification in linear fractional order systems using coloured measurement noise

In the present paper, the identification and estimation problem of a single-input–single-output (SISO) fractional order state-space system will be addressed. A SISO state-space model is considered in which parameters and also state variables should be estimated. The canonical fractional order state-space system will be transformed into a regression equation by using a linear transformation and a shift operator that are appropriate for identification. The identification method provided in this paper is based on a recursive identification algorithm that has the capability of identifying the parameters of fractional order state-space system recursively. Another subject that will be addressed in this paper is a novel fractional order Kalman filter suitable for the systems with coloured measurement noise. The promising performance of the proposed methods is verified using two stable fractional order systems.