基于kernel K-means算法的城市交通客流量分析

基于核函数这种基于统计学习理论的方法,介绍了kernel K-means算法的基本原理与步骤,与传统的K-means算法进行了对比分析,无论是运算速度还是算法有效性,kernel K-means算法都优于传统的K-means算法,并应用于实际的城市交通客流量数据分析实验,结果验证了方法的有效性,为城市交通规律分析、城市规划与交通政策的制定提供了依据。

Local Kernel Dimension Reduction in Approximate Bayesian Computation

Approximate Bayesian Computation (ABC) is a popular sampling method in applications involving intractable likelihood functions. Instead of evaluating the likelihood function, ABC approximates the posterior distribution by a set of accepted samples which are simulated from a generating model. Simulated samples are accepted if the distances between the samples and the observation are smaller than some threshold. The distance is calculated in terms of summary statistics. This paper proposes Local Gradient Kernel Dimension Reduction (LGKDR) to construct low dimensional summary statistics for ABC. The proposed method identifies a sufficient subspace of the original summary statistics by implicitly considering all non-linear transforms therein, and a weighting kernel is used for the concentration of the projections. No strong assumptions are made on the marginal distributions, nor the regression models, permitting usage in a wide range of applications. Experiments are done with simple rejection ABC and sequential Monte Carlo ABC methods. Results are reported as competitive in the former and substantially better in the latter cases in which Monte Carlo errors are compressed as much as possible.

基于函数型数据分析和k-means算法的电力用户分类（英文）

为了对大量电力用户的稀疏、不规律的日耗电量数据进行特征分析,并对用户进行分类,文章提出一种函数性数据聚类分析方法。首先,应用kernel方法将离散的电量数据还原成连续曲线;然后,受Sobolev空间距离的启发,定义了新的函数距离,用于k-means算法进行聚类。以某城市10 000户居民538天的实际用电数据进行实验,得到了用户在不同距离和聚类个数下的聚类原型。实验结果显示,由于选取的用户主要是城市居民,其用电模式比较相似:大高峰时段主要在6—9月,小高峰时段主要在1—2月,日消耗波动较小。而不同用户类别的主要区别体现在用电量的范围上:低耗电用户整体低于13 k W?h/天,高耗电用户接近100 k W?h/天。

求解大规模谱聚类的近似加权核k-means算法

谱聚类将聚类问题转化成图划分问题,是一种基于代数图论的聚类方法.在求解图划分目标函数时,一般利用Rayleigh熵的性质,通过计算Laplacian矩阵的特征向量将原始数据点映射到一个低维的特征空间中,再进行聚类.然而在谱聚类过程中,存储相似矩阵的空间复杂度是O(n2),对Laplacian矩阵特征分解的时间复杂度一般为O(n3),这样的复杂度在处理大规模数据时是无法接受的.理论证明,Normalized Cut图聚类与加权核k-means都等价于矩阵迹的最大化问题.因此,可以用加权核k-means算法来优化Normalized Cut的目标函数,这就避免了对Laplacian矩阵特征分解.不过,加权核k-means算法需要计算核矩阵,其空间复杂度依然是O(n2).为了应对这一挑战,提出近似加权核k-means算法,仅使用核矩阵的一部分来求解大数据的谱聚类问题.理论分析和实验对比表明,近似加权核k-means的聚类表现与加权核k-means算法是相似的,但是极大地减小了时间和空间复杂性.

An Approximation Algorithm Based on Seeding Algorithm for Fuzzy k-Means Problem with Penalties

As a classic NP-hard problem in machine learning and computational geometry,the k-means problem aims to partition the given dataset into k clusters according to the minimal squared Euclidean distance.Different from k-means problem and most of its variants,fuzzy k-means problem belongs to the soft clustering problem,where each given data point has relationship to every center point.Compared to fuzzy k-means problem,fuzzy k-means problem with penalties allows that some data points need not be clustered instead of being paid penalties.In this paper,we propose an O(αk In k)-approximation algorithm based on seeding algorithm for fuzzy k-means problem with penalties,whereαinvolves the ratio of the maximal penalty value to the minimal one.Furthermore,we implement numerical experiments to show the effectiveness of our algorithm.

HMRF半监督近似核k-means算法

信息技术的发展催生了海量数据。聚类有助于发现数据的内在联系,从中挖掘有价值的信息。在对数据进行分析时,容易获得一些关于数据的背景知识,使用这些有限的先验信息指导聚类,可以显著改善聚类的结果。基于隐马尔可夫随机场(Hidden Markov Random Fields,HMRF)的半监督聚类使用成对约束作为监督信息,虽然在很多应用场景中有较好的聚类效果,但是其时间和空间复杂度很高,无法满足大规模数据处理的需要。针对该问题,文中首先分析了HMRF半监督聚类与核k-means的数学联系,使用矩阵的迹将两者的目标函数统一起来;然后,为了降低HMRF半监督聚类的复杂度,提出HMRF半监督近似核k-means算法(HMRF semi-supervised Approximate Kernel K-Means,HMRF-AKKM),通过采样构造近似核矩阵,使用近似核k-means优化聚类的目标函数;最后,在基准数据集上将HMRF-AKKM算法与相关的聚类算法进行对比,分析不同算法在实验中的聚类表现。实验结果表明,在相同的聚类任务上,HMRF-AKKM算法与原始的HMRF半监督聚类具有类似的聚类质量,但是HMRF-AKKM算法的聚类时间更短,说明HMRF-AKKM算法继承了HMRF半监督聚类与近似核k-means的优点。该算法一方面可以充分利用成对约束信息改善聚类质量,另一方面通过采样和矩阵近似提高了聚类效率,而且聚类质量和聚类效率可以通过调节采样比例和成对约束数量来平衡。因此,所提出的HMRF-AKKM算法具有良好的可扩展性,适合处理大规模非线性数据的聚类问题。

An Improved Kernel K-Mean Cluster Method and Its Application in Fault Diagnosis of Roller Bearing

For the kernel K-mean cluster method is run in an implicit feature space, the initial and iterative cluster centers cannot be defined explicitly. Against the deficiency of the initial cluster centers selected in the original space discretionarily in the existing methods, this paper proposes a new method for ensuring the clustering center that virtual clustering centers are defined in the feature space by the original classification as the initial cluster centers and the iteration clustering centers are ensured by the further virtual classification. The improved method is used for fault diagnosis of roller bearing that achieves a good cluster and diagnosis result, which demonstrates the effectiveness of the proposed method.

WAVELET KERNEL SUPPORT VECTOR MACHINES FOR SPARSE APPROXIMATION

Wavelet, a powerful tool for signal processing, can be used to approximate the target func-tion. For enhancing the sparse property of wavelet approximation, a new algorithm was proposed by using wavelet kernel Support Vector Machines (SVM), which can converge to minimum error with bet-ter sparsity. Here, wavelet functions would be firstly used to construct the admitted kernel for SVM according to Mercy theory; then new SVM with this kernel can be used to approximate the target fun-citon with better sparsity than wavelet approxiamtion itself. The results obtained by our simulation ex-periment show the feasibility and validity of wavelet kernel support vector machines.

Approximating and learning by Lipschitz kernel on the sphere

This paper investigates some approximation properties and learning rates of Lipschitz kernel on the sphere. A perfect convergence rate on the shifts of Lipschitz kernel on the sphere, which is faster than O（n-1/2）, is obtained, where n is the number of parameters needed in the approximation. By means of the approximation, a learning rate of regularized least square algorithm with the Lipschitz kernel on the sphere is also deduced.

APPROXIMATION RATES OF ERROR DISTRIBUTION OF DOUBLE KERNEL ESTIMATES OF CONDITIONAL DENSITY

In this paper, the normal approximation rate and the random weighting approximation rate of error distribution of the kernel estimator of conditional density function f(y|x) are studied. The results may be used to construct the confidence interval of f(y|x) .

Wave equation tomographic velocity inversion method based on the Born/Rytov approximation

This paper discusses Born/Rytov approximation tomographic velocity inversion methods constrained by the Fresnel zone. Calculations of the sensitivity kernel function and traveltime residuals are critical in tomographic velocity inversion. Based on the Bom/Rytov approximation of the frequency-domain wave equation, we derive the traveltime sensitivity kemels of the wave equation on the band-limited wave field and simultaneously obtain the traveltime residuals based on the Rytov approximation. In contrast to single-ray tomography, the modified velocity inversion method improves the inversion stability. Tests of the near- surface velocity model and field data prove that the proposed method has higher accuracy and Computational efficiency than ray theory tomography and full waveform inversion methods.

分布式稀疏软大间隔聚类

虽然软大间隔聚类(Soft large margin clustering,SLMC)相比其他诸如K-Means等算法具有更优的聚类性能与某种程度的可解释性,然而当面对大规模分布存储数据时,均遭遇了同样的可扩展瓶颈,其涉及的核矩阵计算需要高昂的时间代价。消减此代价的有效策略之一是采用随机Fourier特征变换逼近核函数,而逼近精度所依赖的特征维度常常过高,隐含着可能过拟合的风险。本文将稀疏性嵌入核SLMC,结合交替方向乘子法(Alternating direction method of multipliers,ADMM),给出了一个分布式稀疏软大间隔聚类算法(Distributed sparse SLMC,DS-SLMC)来克服可扩展问题,同时通过稀疏化获得更好的可解释性。

基于加权UMAP和改进BLS的锂电池温度预测

锂电池热过程的温度预测对锂电池的寿命管理和使用安全有着重要意义。一般电池管理系统热管理依赖准确的热过程模型。然而锂电池热过程的机理复杂,属于强非线性分布参数系统,具有参数时空耦合、时变、强非线性的特点,常规方法难以实现其热过程的精确建模。针对上述问题,提出了一种基于加权UMAP和改进BLS的三段式锂电池热过程建模方式。首先通过引入加权改进的均匀流形逼近与投影(weighted uniform manifold approximation and projection,WUMAP)降维算法解决非线性降维难题的同时保留了数据的全局与局部信息。然后利用一段宽度学习系统(broad learning system,BLS)模型对降维得到的时序数据预测。最后再通过一段粒子群算法优化的混合核宽度学习系统(particle swarm optimization-mixed kernel broad learning system,PSO-MKBLS)模型对时空域温度数据重构。为验证模型有效性,使用平板式32 Ah的Li(Ni_(0.5)Co_(0.2)Mn_(0.3))O_(2)三元软包锂电池的热过程建模试验。实验结果表明:最终模型与改进前相比,R2提高0.0546,MAE和RMSE分别降低0.0082和0.0092;同时与多个对比模型相比,相对误差ARE较低(在0.035以内),并且各误差指标也更好,证明模型具有良好的预测精度。

Local search yields a PTAS for fixed-dimensional k-means problem with penalties

We study a problem called the k-means problem with penalties(k-MPWP),which is a natural generalization of the typical k-means problem.In this problem,we have a set D of client points in R^(d),a set F of possible centers in R^(d),and a penalty cost Pj>O for each point j∈D.We are also given an integer k which is the size of the center point set.We want to find a center point set S■F with size k,choose a penalized subset of clients P■D,and assign every client in D\P to its open center.Our goal is to minimize the sum of the squared distances between every point in D\P to its assigned centre point and the sum of the penalty costs for all clients in P.By using the multi-swap local search technique and under the fixed-dimensional Euclidean space setting,we present a polynomial-time approximation scheme(PTAS)for the k-MPWP.

THE SPARSE REPRESENTATION RELATED WITH FRACTIONAL HEAT EQUATIONS

This study introduces a pre-orthogonal adaptive Fourier decomposition(POAFD)to obtain approximations and numerical solutions to the fractional Laplacian initial value problem and the extension problem of Caffarelli and Silvestre(generalized Poisson equation).As a first step,the method expands the initial data function into a sparse series of the fundamental solutions with fast convergence,and,as a second step,makes use of the semigroup or the reproducing kernel property of each of the expanding entries.Experiments show the effectiveness and efficiency of the proposed series solutions.

Exploring Motor Imagery EEG: Enhanced EEG Microstate Analysis with GMD-Driven Density Canopy Method

The analysis of microstates in EEG signals is a crucial technique for understanding the spatiotemporal dynamics of brain electrical activity.Traditional methods such as Atomic Agglomerative Hierarchical Clustering(AAHC),K-means clustering,Principal Component Analysis(PCA),and Independent Component Analysis(ICA)are limited by a fixed number of microstate maps and insufficient capability in cross-task feature extraction.Tackling these limitations,this study introduces a Global Map Dissimilarity(GMD)-driven density canopy K-means clustering algorithm.This innovative approach autonomously determines the optimal number of EEG microstate topographies and employs Gaussian kernel density estimation alongside the GMD index for dynamic modeling of EEG data.Utilizing this advanced algorithm,the study analyzes the Motor Imagery(MI)dataset from the GigaScience database,GigaDB.The findings reveal six distinct microstates during actual right-hand movement and five microstates across other task conditions,with microstate C showing superior performance in all task states.During imagined movement,microstate A was significantly enhanced.Comparison with existing algorithms indicates a significant improvement in clustering performance by the refined method,with an average Calinski-Harabasz Index(CHI)of 35517.29 and a Davis-Bouldin Index(DBI)average of 2.57.Furthermore,an information-theoretical analysis of the microstate sequences suggests that imagined movement exhibits higher complexity and disorder than actual movement.By utilizing the extracted microstate sequence parameters as features,the improved algorithm achieved a classification accuracy of 98.41%in EEG signal categorization for motor imagery.A performance of 78.183%accuracy was achieved in a four-class motor imagery task on the BCI-IV-2a dataset.These results demonstrate the potential of the advanced algorithm in microstate analysis,offering a more effective tool for a deeper understanding of the spatiotemporal features of EEG signals.

BEST APPROXIMATION FOR WEIERSTRASS TRANSFORM CONNECTED WITH SPHERICAL MEAN OPERATOR

Using reproducing kernels for Hilbert spaces, we give best approximation for Weierstrass transform associated with spherical mean operator. Also, estimates of extremal functions are checked.

A Fast and Effective Multiple Kernel Clustering Method on Incomplete Data

Multiple kernel clustering is an unsupervised data analysis method that has been used in various scenarios where data is easy to be collected but hard to be labeled.However,multiple kernel clustering for incomplete data is a critical yet challenging task.Although the existing absent multiple kernel clustering methods have achieved remarkable performance on this task,they may fail when data has a high value-missing rate,and they may easily fall into a local optimum.To address these problems,in this paper,we propose an absent multiple kernel clustering(AMKC)method on incomplete data.The AMKC method rst clusters the initialized incomplete data.Then,it constructs a new multiple-kernel-based data space,referred to as K-space,from multiple sources to learn kernel combination coefcients.Finally,it seamlessly integrates an incomplete-kernel-imputation objective,a multiple-kernel-learning objective,and a kernel-clustering objective in order to achieve absent multiple kernel clustering.The three stages in this process are carried out simultaneously until the convergence condition is met.Experiments on six datasets with various characteristics demonstrate that the kernel imputation and clustering performance of the proposed method is signicantly better than state-of-the-art competitors.Meanwhile,the proposed method gains fast convergence speed.

Using reproducing kernel for solving a class of partial differential equation with variable-coefficients

How to solve the partial differential equation has been attached importance to by all kinds of fields. The exact solution to a class of partial differential equation with variable-coefficient is obtained in reproducing kernel space. For getting the approximate solution, give an iterative method, convergence of the iterative method is proved. The numerical example shows that our method is effective and good practicability.

C0 Approximation on the Spatially Homogeneous Boltzmann Equation for Maxwellian Molecules

In this paper we study the viscosity analysis of the spatially homogeneous Boltzmann equation for Maxwellian molecules. We first show that the global existence in time of the mild solution of the viscosity equation . We then study the asymptotic behaviour of the mild solution as the coefficients , and an estimate on is derived.