基于特征选择和梯度提升算法的高光谱遥感地物识别

高光谱遥感数据波段数目较多,且波段之间的相关性高,影响到敏感波段在地物识别中的作用,并造成大量冗余计算,降低时效.提出了一种随机森林结合递归特征消除的敏感特征选择方案,以提高高光谱遥感地物识别的精度与效率.通过RF-RFE特征选择方法得到最优特征组合,并运用LightGBM和XGBoost等提升算法来提高分类精度.在江苏省常州的茶树数据集上进行分类实验时,在原始数据上的分类精度达到了94.27%和94.45%;在特征选择出的最优特征子集上进行实验时,分类精度达到了94.40%和94.36%.实验结果表明,该方案的分类精度要优于决策树和朴素贝叶斯等传统分类算法,同时大幅减少了运算量,取得了较好的识别效果,具有一定的推广和应用价值.     

基于BSA-ELM模型的建筑工程施工成本预测研究

针对建筑工程施工成本管理中成本难以预测的问题,提出用鸟群算法(BSA)优化极限学习机(ELM)模型的参数.首先,利用BSA对ELM模型的输入权值和偏置值进行优化;其次,构建出BSA-ELM建筑工程施工成本预测模型;最后,将BSA-ELM模型与实际工程施工成本数据相结合进行验证.结果表明:模型在成本预测中的精度比ELM模型、CSO-ELM模型、PSO-ELM模型和BP神经网络模型预测精度高,也为类似预测问题提供了一种新的预测方法.     

基于指数Laplace损失函数的回归估计鲁棒超限学习机

实际问题的数据集通常受到各种噪声的影响,超限学习机(extreme learning machine, ELM)对这类数据集进行学习时,表现出预测精度低、预测结果波动大.为了克服该缺陷,采用了能够削弱噪声影响的指数Laplace损失函数.该损失函数是建立在Gauss核函数基础上,具有可微、非凸、有界且能够趋近于Laplace函数的特点.将其引入到超限学习机中,提出了鲁棒超限学习机回归估计(exponential Laplace loss function based robust ELM for regression, ELRELM)模型.利用迭代重赋权算法求解模型的优化问题.在每次迭代中,噪声样本点被赋予较小的权值,能够有效地提高预测精度.真实数据集实验验证了所提出的模型相比较于对比算法具有更优的学习性能和鲁棒性.     

基于DE-ELM的电池SOC预测研究

针对极限学习机的随机性较大的问题,提出一种基于差分演化的极限学习机算法模型(DE-ELM).采用差分演化算法(DE)对极限学习机(ELM)随机给定的输入权值矩阵和隐含层阈值进行寻优,降低了随机性给ELM造成的影响,减少ELM网络震荡,提高了ELM预测精度.并且将DE-ELM应用在电池SOC的预测上,同时与ELM和BP神经网络的预测进行了对比,结果表明:DE-ELM在电池SOC预测上的表现优于ELM和BP神经网络,能满足电池SOC的预测精度要求.     

基于模糊c均值算法和人工蜂群算法的无监督波段选择

波段选择是高光谱影像处理中一种重要的降维方法.在类标签不可获得的情况下,如何选择出一个具有代表性的波段子集是一个挑战性的问题.为了解决高光谱数据维数灾难以及光谱空间冗余的问题,基于模糊C均值算法(Fuzzy c-means,FCM),人工蜂群算法(Artificial Bee Colony, ABC)与极大熵准则(Maximum Entropy, ME),文章提出了一种新的无监督波段选择方法.该方法首先通过FCM算法将相似的波段划分到一个波段子集中,然后以ME为ABC算法中的适应度函数,寻找优化的波段子集.为验证该算法的有效性,在三个典型的高光谱数据集上,将所提出的方法和其它一些有效的波段选择算法进行了分类精度和计算时间对比.实验结果表明,所提出的算法不但可以得到高的分类精度,同时在计算时间上也具有明显的优势.     

数据空间结构性及在KNN算法中的应用

数据空间结构性是多维数据客观存在的本征特征,是数据挖掘的重要内容.通过统计学的方法,分析多维数据空间属性变量和类型变量之间的结构特征,可以准确刻画数据在多维变量空间的相关性及其各向异性.数据空间结构特征可以用于机器学习算法的改进和提升,以提高模式识别的效果.融合了数据空间结构特征的KNN算法在稳定性和识别精度上均优于传统算法.通过在苏里格气田苏东41-33区块复杂碳酸盐岩的岩性识别中的应用表明,与传统KNN算法相比,数据空间结构的引入能提高识别准确率12.35%,并显示出算法的灵活性和适用性.多维数据空间结构的研究对机器学习算法的泛化能力和迁移性的提升等方面具有促进作用.     

基于MCMC主成分回归分析方法的经济增长因素研究

将MCMC算法融合到主成分回归分析模型中,提出MCMC主成分回归分析方法.新方法既具有有效避免解释变量之间的多重共线性问题以及简化回归方程结构的主成分回归分析方法的优势,又能够充分利用MCMC算法的融合先验信息、模型信息及样本似然函数的长处.将方法应用于对嘉兴市1997年至201.0年的经济发展指标的数据建模分析,结果表明,方法能有效克服现有分析方法的不足,建立预测精度更高的模型.     

基于Adaboost和正则化ELM的混合金融时间序列预测模型及其应用

为提高金融时间序列的预测精度,本文提出了基于MODWT、MCP变量选择方法和RELM_Adaboost的混合预测模型。该模型由三步构成:第一步,收集特征变量,包括MODWT分解得到的特征变量以及常用的技术指标;第二步,利用MCP惩罚方法从上述特征变量中选取重要的作为输入变量;第三步,利用Mnet惩罚正则化ELM,将RELM视作弱预测器,然后用Adaboost算法生成强预测器进行预测。实证结果显示:第一,经过MCP方法的筛选,最终的输入变量中不仅包含常用技术指标,还有小波分解所得的变量。第二,混合预测模型RELM_Adaboost有良好的泛化误差表现。本文提出的模型在量化交易时代具有良好的应用前景。     

基于限制玻尔兹曼机的极限学习机方法

针对高维数据中存在冗余以及极限学习机(ELM)存在随机给定权值导致算法性能不稳定等问题,将限制玻尔兹曼机(RBM)与ELM相结合提出了基于限制玻尔兹曼机优化的极限学习机算法(RBM-ELM).通过限制玻尔兹曼机对原始数据进行特征降维的同时,得到ELM输入层权值和隐含层偏置的优化参数.实验结果表明,相比较随机森林,逻辑回归,支持向量机和极限学习机四种机器学习算法,RBM-ELM算法能获得较高的分类精度.     

偏最小二乘建模在R软件中的实现及实证分析

通过介绍偏最小二乘（PLS）的建模和显著性检验原理,解决了小样本多变量且变量间存在多重共线性的回归问题,建立了多变量对多变量的回归模型,并使用R软件（版本为Ri3862．15．1）实现了PLS建模;最后基于葡萄和葡萄酒理化指标数据进行了实证分析．     

两类变时间步长的非线性Galerkin算法的稳定性

1．引言近年来,随着计算机的飞速发展,人们越来越关心非线性发展方程解的渐进行为．为了较精确地描述解在时间t→∞时的渐进行为,人们发展了一类惯性算法,即非线性Galerkin算法．该算法是将来解空间分解为低维部分和高维部分,相应的方程可以分别投影到它们上面,它的解也相应地分解为两部分,大涡分量和小涡分量;然后核算法给出大涡分量和小涡分量之间依赖关系的一种近似,以便容易求出相应的近似解．许多研究表明,非线性Galerkin算法比通常的Galerkin算法节省可观的计算量．当数值求解微分方程时,计算机只能对已知数据进行有限位…     

两个带变时间步的两重网格算法的稳定性分析

对用于求解非线性发展方程的两个带变时间步的两重网格算法，对空间变量用有限元离散，对时间变量分别用一阶精度Euler显式和二阶精度半隐式差分格式离散，然后构造两重网格算法，通过深入的稳定性分析，得出本文的算法优于标准全离散有限元算法。     

A kernel extreme learning machine algorithm based on improved particle swam optimization

Kernel extreme learning machine (KELM) increases the robustness of extreme learning machine (ELM) by turning linearly non-separable data in a low dimensional space into a linearly separable one. However, the internal power parameters of ELM are initialized at random, causing the algorithm to be unstable. In this paper, we use the active operators particle swam optimization algorithm (APSO) to obtain an optimal set of initial parameters for KELM, thus creating an optimal KELM classifier named as APSO-KELM. Experiments on standard genetic datasets show that APSO-KELM has higher classification accuracy when being compared to the existing ELM, KELM, and these algorithms combining PSO/APSO with ELM/KELM, such as PSO-KELM, APSO-ELM, PSO-ELM, etc. Moreover, APSO-KELM has good stability and convergence, and is shown to be a reliable and effective classification algorithm.     

On the convergence of the partial least squares path modeling algorithm

This paper adds to an important aspect of Partial Least Squares (PLS) path modeling, namely the convergence of the iterative PLS path modeling algorithm. Whilst conventional wisdom says that PLS always converges in practice, there is no formal proof for path models with more than two blocks of manifest variables. This paper presents six cases of non-convergence of the PLS path modeling algorithm. These cases were estimated using Mode A combined with the factorial scheme or the path weighting scheme, which are two popular options of the algorithm. As a conclusion, efforts to come to a proof of convergence under these schemes can be abandoned, and users of PLS should triangulate their estimation results.     

基于精英区域学习的多种群自适应的差分进化算法

为了进一步提高差分进化算法的收敛速度、算法精度和稳定性,采用多种群技术来增加算法收敛速度和降低复杂度;利用精英区域学习策略来对算法的全局搜索能力和算法精度进一步提升,引进自适应免疫搜索策略,以实现自适应修正差分算法的变异因子和交叉因子。通过五个测试函数,把本文算法与最新文献中的算法进行对比,表明算法在收敛速度、精度和高维问题寻优能力方面的优越性。     

基于熵权TOPSIS-PSO-ELM的制造企业绿色创新能力评价模型及实证研究

为客观和准确地评价制造企业绿色创新能力,本文构建了制造企业绿色创新能力评价指标体系,提出了基于熵权TOPSIS的粒子群(PSO)优化极限学习机(ELM)集成学习算法的制造企业绿色创新能力评价模型。首先运用熵权法客观确定指标权重,结合TOPSIS测度并综合评价制造企业绿色创新能力,然后将评价值作为先验样本进行极限学习机的训练与测试,训练过程中利用PSO优化极限学习机的网络结构与连接权值,从而对绿色创新能力进行全面的分析和评价。最后以60家制造企业为例进行实证分析,并将熵权TOPSIS-PSO-ELM算法与极限学习机回归拟合算法对比,结果表明:基于熵权TOPSIS-PSO-ELM模型所得评价结果较已有方法更为准确可靠。此外,为进一步提高我国制造企业绿色创新发展能力提出了理论建议。     

Determination of the relaxation and retardation spectra – a view from the up-to-date signal processing perspective

The problem of determination of relaxation and retardation spectra (RRS) is considered from the viewpoint of up-to-date signal processing. It is shown that the recovery of RRS represents the Mellin deconvolution problem, which transforms into the Fourier deconvolution problem for data on a logarithmic time or frequency scale, where it can also be treated as the inverse filtering problem. On this basis, discrete deconvolution (inverse) filters operating with geometrically sampled data are proposed to use as RRS estimators. Appropriate frequency responses and algorithms are derived for estimating RRS from eight different material functions. The noise amplification coefficient is suggested to use as a measure for quantifying the degree of ill-posedness and illconditioness of the RRS recovery problem and algorithms. A methodology is developed for designing RRS estimators with a desired noise amplification, producing maximum accurate spectra for available limited input data. Practical algorithms for determining RRS are proposed, and their performance is studied. The algorithms suggested are compared with the so-called moving-average formulae. It is demonstrated that the minimum frequency range for recovering the point estimate of a relaxation spectrum depends on the allowable noise amplification (the degree of ill-conditioness) and is in no way limited by 1.36 decades, as it is stated by the sampling localization theorem.     

An algorithm of sequential systems of linear equations for nonlinear optimization problems with arbitrary initial point

For current sequential quadratic programming (SQP) type algorithms, there exist two problems: (i) in order to obtain a search direction, one must solve one or more quadratic programming subproblems per iteration, and the computation amount of this algorithm is very large. So they are not suitable for the large-scale problems; (ii) the SQP algorithms require that the related quadratic programming subproblems be solvable per iteration, but it is difficult to be satisfied. By using ε-active set procedure with a special penalty function as the merit function, a new algorithm of sequential systems of linear equations for general nonlinear optimization problems with arbitrary initial point is presented. This new algorithm only needs to solve three systems of linear equations having the same coefficient matrix per iteration, and has global convergence and local superlinear convergence. To some extent, the new algorithm can overcome the shortcomings of the SQP algorithms mentioned above. Project partly supported by the National Natural Science Foundation of China and Tianyuan Foundation of China.     

Extreme logistic regression

Kernel logistic regression (KLR) is a very powerful algorithm that has been shown to be very competitive with many state-of the art machine learning algorithms such as support vector machines (SVM). Unlike SVM, KLR can be easily extended to multi-class problems and produces class posterior probability estimates making it very useful for many real world applications. However, the training of KLR using gradient based methods or iterative re-weighted least squares can be unbearably slow for large datasets. Coupled with poor conditioning and parameter tuning, training KLR can quickly design matrix become infeasible for some real datasets. The goal of this paper is to present simple, fast, scalable, and efficient algorithms for learning KLR. First, based on a simple approximation of the logistic function, a least square algorithm for KLR is derived that avoids the iterative tuning of gradient based methods. Second, inspired by the extreme learning machine (ELM) theory, an explicit feature space is constructed through a generalized single hidden layer feedforward network and used for training iterative re-weighted least squares KLR (IRLS-KLR) and the newly proposed least squares KLR (LS-KLR). Finally, for large-scale and/or poorly conditioned problems, a robust and efficient preconditioned learning technique is proposed for learning the algorithms presented in the paper. Numerical results on a series of artificial and 12 real bench-mark datasets show first that LS-KLR compares favorable with SVM and traditional IRLS-KLR in terms of accuracy and learning speed. Second, the extension of ELM to KLR results in simple, scalable and very fast algorithms with comparable generalization performance to their original versions. Finally, the introduced preconditioned learning method can significantly increase the learning speed of IRLS-KLR.     

A multi-start algorithm for a balanced real-world Open Vehicle Routing Problem

The aim of this paper is to solve a real-world problem proposed by an international company operating in Spain and modeled as a variant of the Open Vehicle Routing Problem in which the makespan, i.e., the maximum time spent on the vehicle by one person, must be minimized. A competitive multi-start algorithm, able to obtain high quality solutions within reasonable computing time is proposed. The effectiveness of the algorithm is analyzed through computational testing on a set of 19 school-bus routing benchmark problems from the literature, and on 9 hard real-world problem instances.