Support vector machine via nonlinear rescaling method

In this paper we construct the linear support vector machine (SVM) based on the nonlinear rescaling (NR) methodology (see [Polyak in Math Program 54:177–222, 1992; Polyak in Math Program Ser A 92:197–235, 2002; Polyak and Teboulle in Math Program 76:265–284, 1997] and references therein). The formulation of the linear SVM based on the NR method leads to an algorithm which reduces the number of support vectors without compromising the classification performance compared to the linear soft-margin SVM formulation. The NR algorithm computes both the primal and the dual approximation at each step. The dual variables associated with the given data-set provide important information about each data point and play the key role in selecting the set of support vectors. Experimental results on ten benchmark classification problems show that the NR formulation is feasible. The quality of discrimination, in most instances, is comparable to the linear soft-margin SVM while the number of support vectors in several instances were substantially reduced.     

Difference schemes on uniform grids performed by general discrete operators

Our aim is to set the foundations of a discrete vectorial calculus on uniform n-dimensional grids, that can be easily reformulated on general irregular grids. As a key tool we first introduce the notion of tangent space to any grid node. Then we define the concepts of vector field, field of matrices and inner products on the space of grid functions and on the space of vector fields, mimicking the continuous setting. This allows us to obtain the discrete analogous of the basic first order differential operators, gradient and divergence, whose composition define the fundamental second order difference operator. As an application, we show that all difference schemes, with constant coefficients, for first and second order differential operators with constant coefficients can be seen as difference operators of the form for suitable choices of q, and  . In addition, we characterize special properties of the difference scheme, such as consistency, symmetry and positivity in terms of q, and  .     

Space–time integrated least squares: a time‐marching approach

A time‐marching formulation is derived from the space–time integrated least squares (STILS) method for solving a pure hyperbolic convection equation and is numerically compared to various known methods. Copyright © 2004 John Wiley & Sons, Ltd.     

基于Karhunen-Loève变换和小波谱特征矢量量化的三维谱像数据压缩

提出了基于Karhunen Lo埁ｖｅ变换的小波谱特征矢量量化三维谱像数据压缩方法耍幔颍瑁酰睿澹?Lo埁ｖｅ变换 /小波变换 /小波谱特征矢量量化方法应用了Karhunen Lo埁ｖｅ变换的消除谱相关性优良性能 ,应用二维小波变换消除空间相关性 ,在小波变换域内应用二维集分割嵌入块编码和一维谱特征矢量量化对三维谱像数据压缩 ,获得较高的压缩性能。实验结果表明 :Karhunen Lo埁ｖｅ变换 /小波变换 /小波谱特征矢量量化编码比Karhunen Lo埁ｖｅ变换 /小波变换 /改进对块零树编码和Karhunen Lo埁ｖｅ变换 /小波变换 /快速矢量量化编码方法在同样压缩比条件下 ,峰值信噪比提高 2dB和 1dB以上 ,而速度提高了 1.5和 8倍 ,整体压缩性能有较大的提高     

Particle Size Analysis by Transmission Fluctuation Spectrometry: Experimental Results Obtained with a Gaussian Beam and Analog Signal Processing

Transmission fluctuation spectrometry (TFS) is a method for the analysis of particle size distributions based on the statistical fluctuations of a transmission signal. Complete information on the PSD and particle concentration can be retrieved by a special transformation of the transmission signal, whereby the expectancy of the transmission square (ETS) is determined after the signal has been subjected to a procedure of spatial and temporal averaging. By varying the averaging parameters over a wide range, a spectrum of ETSs is obtained and introduced into a linear equation system, which yields the PSD. In the experimental realization presented here, variable temporal averaging is realized in the frequency domain with a series of low pass filters at different cutoff frequencies while spatial averaging inevitably occurs as the particles pass through a focused Gaussian beam of finite cross section. Experimental results on spherical particles (glass beads) and non‐spherical particles (SiC) are presented. The PSDs are resolved in 30 intervals within a particle size range from 1–1000 μm, employing a modified Chahine inversion algorithm. So far, the measurements are limited to moderate particle concentrations. Some influences affecting the measurements, especially for higher particle concentrations, are discussed in detail.     

A new type SVM-projected SVM

Support vector machine (SVM), developed by Vapnik et al., is a new and promising technique for classification and regression and has been proved to be competitive with the best available learning machines in many applications. However, the classification speed of SVM is substantially slower than that of other techniques with similar generalization ability. A new type SVM named projected SVM (PSVM), which is a combination of feature vector selection (FVS) method and linear SVM (LSVM), is proposed in present paper. In PSVM, the FVS method is first used to select a relevant subset (feature vectors, FVs) from the training data, and then both the training data and the test data are projected into the subspace constructed by FVs, and finally linear SVM(LSVM) is applied to classify the projected data. The time required by PSVM to calculate the class of new samples is proportional to the count of FVs. In most cases, the count of FVs is smaller than that of support vectors (SVs), and therefore PSVM is faster than SVM in running. Compared with other speeding-up techniques of SVM, PSVM is proved to possess not only speeding-up ability but also de-noising ability for high-noised data, and is found to be of potential use in mechanical fault pattern recognition.     

Robust Adaptive Identification of Fuzzy Systems with Uncertain Data

This study presents a method of adaptive identification of parameters describing Sugeno fuzzy inference system in presence of bounded disturbances while maintaining the readability and interpretability of the fuzzy model during and after identification. This method do not require any a priori knowledge of a bound on the disturbance and noise and of a bound on the unknown parameters values. The method can be used for the robust and adaptive identification of slowly time varying nonlinear systems using fuzzy inference systems. The suggested method was used to build a fuzzy expert system that approximates the functional relationship between physical fitness and some of the measurable physiological parameters by their real measurements and opinion (human-experiences) of a medical expert.     

广义向量均衡问题的解与解集的凸性

引进集值映射的锥真拟凹概念，讨论一类具有集值映射的广义向量衡问题解的存在性与解集的凸性。     

Reagent concentration effects in the REM resin solid phase synthesis of tertiary amines

The use of reagent concentration has resulted in increased rates for all stages of the REM resin synthesis of tertiary amines. These increases in rate translate into faster reaction times, higher yields and lower reagent consumption. Of the methods examined, the most successful was the use of perfluorous solvents, either alone or with a small amount of organic co-solvent.     

Monotone Mappings and Flows of Viscous Media

We establish solvability of the boundary-value problems describing stationary and periodic flows of viscoplastic media. In the case of stationary flows we study the question of convergence of the Galerkin method. For the problem of periodic flows we prove a version of the second Bogolyubov theorem.