RBF网络的微分进化正交最小二乘算法

研究用于径向基函数（RBF）网络训练的一种微分进化正交最小二乘（DEOLS）算法。把微分进化（DE）算法的种群作为正交最小二乘（OLS）算法的候选径向基函数集合,利用OLS对DE的种群个体进行评断,以确定RBF网络的隐结点的数目、中心和宽度。该算法融合了DE的强大搜索能力和OLS的高效评断能力,隐结点的选择比OLS要合理,同时避免DE的复杂性。最后使用实验验证了该算法的优越性。     

Nonlinear model structure design and construction using orthogonal least squares and D-optimality design

A very efficient learning algorithm for model subset selection is introduced based on a new composite cost function that simultaneously optimizes the model approximation ability and model robustness and adequacy. The derived model parameters are estimated via forward orthogonal least squares, but the model subset selection cost function includes a D-optimality design criterion that maximizes the determinant of the design matrix of the subset to ensure the model robustness, adequacy, and parsimony of the final model. The proposed approach is based on the forward orthogonal least square (OLS) algorithm, such that new D-optimality-based cost function is constructed based on the orthogonalization process to gain computational advantages and hence to maintain the inherent advantage of computational efficiency associated with the conventional forward OLS approach. Illustrative examples are included to demonstrate the effectiveness of the new approach.     

A systematic investigation of geostatistical image fusion for the improvement of the spectral fidelity and spatial detail in Landsat MS imagery

This is the first systematic investigation into the assumptions of image fusion using regression Kriging (RK) – a geostatistical method – illustrated with Landsat MS (multispectral) and SPOT (Satellite Pour l’Observation de la Terre) panchromatic images. The efficiency of different linear regression and Kriging methods in the fusion process is examined by visual and quantitative indicators. Results indicate a trade-off between spectral fidelity and spatial detail preservation for the GLS (generalized least squares regression) and OLS (ordinary least squares regression) methods in the RK process: OLS methods preserve more spatial detail, while GLS methods retain more spectral information from the MS images but at a greater computational cost. Under either OK (ordinary Kriging) or UK (universal Kriging) with either OLS or GLS, the spherical variogram improves spatial details from the panchromatic image, while the exponential variogram maintains more spectral information from the MS image. Overall, RK-based fusion methods outperform conventional fusion approaches from both the spectral and spatial point of view.     

The automatic model selection and variable kernel width for RBF neural networks

The Orthogonal Least Squares (OLS) algorithm has been extensively used in basis selection for RBF networks, but it is unable to perform model selection automatically because the tolerance ρ must be specified manually. This introduces noise and it is difficult to implement in the parametric complexity of real-time system. Therefore, a generic criterion that detects the optimum number of its basis functions is proposed. In this paper, not only the Bayesian Information Criterion (BIC) method, used for fitness calculation, is incorporated into the basis function selection process of the OLS algorithm for assigning its appropriate number, but also a new method is developed to optimize the widths of the Gaussian functions in order to improve the generalization performance. The augmented algorithm is employed to the Radial Basis Function Neural Networks (RBFNN) for known and unknown noise nonlinear dynamic systems and its performance is compared with the standard OLS; experimental results show that both the efficacy of BIC for fitness calculation and the importance of proper choice of basis function widths are significant.     

Scalable-Width Temporal Edge Detection for Recursive Background Recovery in adaptive background modeling

Background modeling is a preliminary processing task that generates background or reference frame for moving object detection. Apart from tracking background scene, a good quality background model will prevent false detection. Fuzzy Running Average (FRA) is an efficient background modeling scheme which employs a Fuzzy Inference System (FIS). Its high selectivity in background update prevents foreground object from appearing in the reference frame. Later, Extended Fuzzy Running Average (EFRA) was developed to allow FRA to recover the occlusion if a background object starts moving. However, the recovery rate of EFRA is limited due to the use of fixed width for detecting the occlusion's edge. In this paper, a newly developed method based on Scalable-Width Temporal Edge Detection (SWTED) is proposed to enhance the EFRA performance in locating and recovering the occlusion with higher rate. The results obtained show that the improved EFRA significantly outperforms FRA in background tracking. The algorithm is also well suited for real-time implementation.     

基于Kalman-OLS的聚丙烯熔融指数软测量

针对聚丙烯装置熔融指数软测量中的非线性和多工况切换操作问题,提出1种基于卡尔曼滤波-正交最小二乘(Kalman-OLS)的非线性自适应软测量方法。通过对聚丙烯装置反应系统进行机理分析,采用正交最小二乘方法(OLS)来拟和辅助变量和熔融指数之间的非线性关系。OLS方法的优化目标函数中同时考虑基于留一法均方误差(LOO MSE)和模型参数的局部正则化(LR),以提高模型的稀疏性和泛化能力。为适应装置多工况操作的现状,进一步提出使用Kalman滤波器对OLS模型参数进行自适应更新。工业数据应用结果表明,提出的Kal-man-OLS方法能够比偏最小二乘(PLS)、OLS方法更准确的预报聚丙烯熔融指数的变化。     

Trimming influential observations for improved single-index model estimated sufficient summary plots

The detection and subsequent treatment of influential observations have been well covered with respect to ordinary least squares (OLS) under an assumed multiple linear regression (MLR) model using measures such as Cook’s Distance. However, OLS can be shown to be a useful method under a much wider variety of models. The purpose of this paper is twofold. Firstly we introduce a new diagnostic, similar to Cook’s Distance, that is useful for detecting influential observations under an assumed single-index model. Secondly we show, via simulation, how trimming observations according to such diagnostics can greatly benefit the analysis even when no gross outliers are evident.     

Combined analytical and FEM methods for parameters calculation of detailed model for dry-type transformer

Non-flammable characteristic of cast-resin dry-type transformers make them suitable for different kind of usages. This paper presents a method of how to obtain high frequency model of these transformers. For this purpose a detailed model is used and parameters of the model including inductances, capacitances and resistances are calculated using FEM (finite elements method). The effect of the frequency is considered in the inductance calculation using FEM. In order to validate the model, a setup was constructed for testing on high-voltage winding of dry-type transformer. The simulation results were compared with the experimental data measured from FRA (frequency response analysis) and impulse tests.     

Joint photometric and geometric image registration in the total least square sense

This paper presents a novel robust image alignment technique that performs joint geometric and photometric registration in the total least square (TLS) sense. Therefore, we employ the total least square metric instead of the ordinary least square (OLS) metric, which is commonly used in the literature. While the OLS model is sufficient to tackle geometric registration problems, it gives no mutually consistent estimates when dealing with photometric deformations. By introducing a new TLS model, we obtain mutually consistent parameters. Experimental results show that our method is indeed more consistent and accurate in presence of noise compared to existing joint registration algorithms.     

Impacts of Interval Computing on Stock Market Variability Forecasting

This study uses the interval computing approach to forecast the annual and quarterly variability of the stock market. We find that the forecasting accuracy is significantly higher than the OLS lower and upper bound forecasting. The strength of the interval computing comes from its data processing. It uses lower and upper bound information simultaneously, no variability information is lost in parameter estimation. The quarterly interval (variability) forecasts suggest that the interval computing method outperforms the OLS lower and upper bound forecasting in both stable and volatile periods.      

基于改进资源分配网络的企业能耗单元输入输出模型研究

针对企业复杂能耗单元输入输出模型研究的要求,研究了基于改进资源分配网络的能耗单元输入输出模型辨识方法。针对常用的资源分配网络存在的问题,提出了一种基于粗糙集和正交最小二乘的资源分配网络设计方法,通过粗糙集数据分析与处理提取训练样本中典型的数据特征,再结合正交最小二乘学习算法选取对输出能量贡献最大的数据中心加入到隐层节点。实例仿真结果表明,采用该方法辨识企业能耗单元输入输出模型具有结构简单、训练快捷、泛化能力较好等优点。     

On the implementation of frontier-to-root tree automata inrecursive neural networks

We explore the node complexity of recursive neural network implementations of frontier-to-root tree automata (FRA). Specifically, we show that an FRAO (Mealy version) with m states, l input-output labels, and maximum rank N can be implemented by a recursive neural network with O( radical(log l+log m)lm(N)/log l+N log m) units and four computational layers, i.e., without counting the input layer. A lower bound is derived which is tight when no restrictions are placed on the number of layers. Moreover, we present a construction with three computational layers having node complexity of O((log l+log m) radicallm (N)) and O((log l+log m)lm(N)) connections. A construction with two computational layers is given that implements any given FRAO with a node complexity of O(lm(N)) and O((log l+log m)lm(N)) connections. As a corollary we also get a new upper bound for the implementation of finite-state automata into recurrent neural networks with three computational layers.     

The mass appraisal of the real estate by computational intelligence

Mass appraisal is the systematic appraisal of groups of properties as of a given date using standardized procedures and statistical testing. Mass appraisal is commonly used to compute real estate tax. There are three traditional real estate valuation methods: the sales comparison approach, income approach, and the cost approach. Mass appraisal models are commonly based on the sales comparison approach. The ordinary least squares (OLS) linear regression is the classical method used to build models in this approach. The method is compared with computational intelligence approaches – support vector machine (SVM) regression, multilayer perceptron (MLP), and a committee of predictors in this paper. All the three predictors are used to build a weighted data-depended committee. A self-organizing map (SOM) generating clusters of value zones is used to obtain the data-dependent aggregation weights. The experimental investigations performed using data cordially provided by the Register center of Lithuania have shown very promising results. The performance of the computational intelligence-based techniques was considerably higher than that obtained using the official real estate models of the Register center. The performance of the committee using the weights based on zones obtained from the SOM was also higher than of that exploiting the real estate value zones provided by the Register center.     

OLS with multiple high dimensional category variables

A new algorithm is proposed for OLS estimation of linear models with multiple high-dimensional category variables. It is a generalization of the within transformation to arbitrary number of category variables. The approach, unlike other fast methods for solving such problems, provides a covariance matrix for the remaining coefficients. The article also sets out a method for solving the resulting sparse system, and the new scheme is shown, by some examples, to be comparable in computational efficiency to other fast methods. The method is also useful for transforming away groups of pure control dummies. A parallelized implementation of the proposed method has been made available as an R-package lfe on CRAN.     

Tobit model analysis of mutations causing haemophilia

Haemophilia A is a genetic disease resulting from deficiency of factor VIII. The database of mutations causing haemophilia A has been developed by the worldwide collaboration. In this study, we predicted the severity of haemophilia A using ordinary least squares regression (OLS) and Tobit model regression in factor VIII. As parameters, we used four physical–chemical parameters of amino acids and the accessible surface area. As the result, there is a difference between OLS and Tobit model analysis in the prediction of haemophilia A severity. We found that the predicted value by Tobit model in severe data is close to the observed value. These results suggest that Tobit model regression analysis is superior to estimating severity of haemophilia A in severe data than OLS.     

模糊预测控制在pH中和过程中的应用

针对pH中和过程,提出了一种基于T-S模型的模糊预测控制算法,以实现系统的滚动优化控制。T-S模糊模型的前件和后件参数分别采用模糊C均值聚类(FCM)和正交最小二乘法(OLS)进行离线或在线辨识。在每一个采样时刻以当前辨识出的T-S模型为基础实现系统的局部动态线性化,再根据线性化模型对pH过程实施广义预测控制(GPC),得到当前的控制量。仿真表明了该控制方法具有较小的超调性质,且在扰动作用下能快速跟踪到设定值,具有很强的鲁棒性。     

Formulation of soil angle of shearing resistance using a hybrid GP and OLS method

In the present study, a prediction model was derived for the effective angle of shearing resistance (?′) of soils using a novel hybrid method coupling genetic programming (GP) and orthogonal least squares algorithm (OLS). The proposed nonlinear model relates ?′ to the basic soil physical properties. A comprehensive experimental database of consolidated-drained triaxial tests was used to develop the model. Traditional GP and least square regression analyses were performed to benchmark the GP/OLS model against classical approaches. Validity of the model was verified using a part of laboratory data that were not involved in the calibration process. The statistical measures of correlation coefficient, root mean squared error, and mean absolute percent error were used to evaluate the performance of the models. Sensitivity and parametric analyses were conducted and discussed. The GP/OLS-based formula precisely estimates the ?′ values for a number of soil samples. The proposed model provides a better prediction performance than the traditional GP and regression models.     

一种新的RBF网络两级学习设计方法

为了简化径向基网络结构,构造出良好泛化性能力的网络,提出了一种径向基(RBF)网络的两级学习新设计方法.在下级将正交最小二乘法(OLS)与A-最优设计方法(A-opt)相结合(OLS+A-opt),引入一种基于A-最优设计准则的混合代价函数,同时优化网络模型的逼近性能及模型的充分性,自动构建结构节俭的RBF网络模型;而方法中的关键学习参数A-最优代价系数通过上级粒子群优化方法(PSO)优化获取最佳值.仿真结果表明该方法所设计的RBF网络不仅具有较好的泛化性能,而且也具有良好的模型鲁棒性及充分性,是一种有效的RBF网络设计方法.     

Land cover in East Asia classified using Terra MODIS and DMSP OLS products

Land cover is classified over East Asia using 250‐m Moderate Resolution Imaging Spectroradiometer (MODIS) land surface reflectance, MODIS snow cover and Operational Linescan System (OLS) human settlement data. The classification method includes a decision tree classification scheme that considers 11 kinds of land surface features derived from the OLS product and the time series of two MODIS products in 2000. The decision tree was defined manually based on the experiment because of insufficient training data, ease of tuning by visual interpretation, and extensibility to further research. The resulting classification is compared to three kinds of reference data, i.e. MODIS land cover product, Chinese digital land cover map, and Chinese census. The land cover classification can be input into a hydrological model applied to the Yellow River in China.