Consistency of regularized spectral clustering

Clustering is a widely used technique in machine learning, however, relatively little research in consistency of clustering algorithms has been done so far. In this paper we investigate the consistency of the regularized spectral clustering algorithm, which has been proposed recently. It provides a natural out-of-sample extension for spectral clustering. The presence of the regularization term makes our situation different from that in previous work. Our approach is mainly an elaborate analysis of a functional named the clustering objective. Moreover, we establish a convergence rate. The rate depends on the approximation property and the capacity of the reproducing kernel Hilbert space measured by covering numbers. Some new methods are exploited for the analysis since the underlying setting is much more complicated than usual. Some new methods are exploited for the analysis since the underlying setting is much more complicated than usual.     

非线性系统的高阶迭代学习控制算法

本文对一般非线性系统,提出了三种高阶迭代学习控制算法:(Ⅰ)u_(h 1)=sum from j=1 to r(P_ju_(k-j 1) Г_je_(h-j 1));(Ⅱ)u_(k 1)=sum from j=1 to r(P_ju_(k-j 1) F_je_(k-j 1));(Ⅲ)u_(k 1)=sum from j=1 to r{P_ju_(k-j 1) (Г_j F_jd/(dt))e_(k-j 1)},其中u_(k 1)=u_(k 1)(t)表示系统第k 1次运行时的输入;e_k=y_k-y_d;y_d是系统所期望的输出;y_k是系统第k次运行时的输出;P_j,Г_j,F_j(j=1,…,r)是常数阵;进而给出了比较弱的收敛性条件。     

Adaptive modeling, adaptive data assimilation and adaptive sampling

For efficient progress, model properties and measurement needs can adapt to oceanic events and interactions as they occur. The combination of models and data via data assimilation can also be adaptive. These adaptive concepts are discussed and exemplified within the context of comprehensive real-time ocean observing and prediction systems. Novel adaptive modeling approaches based on simplified maximum likelihood principles are developed and applied to physical and physical–biogeochemical dynamics. In the regional examples shown, they allow the joint calibration of parameter values and model structures. Adaptable components of the Error Subspace Statistical Estimation (ESSE) system are reviewed and illustrated. Results indicate that error estimates, ensemble sizes, error subspace ranks, covariance tapering parameters and stochastic error models can be calibrated by such quantitative adaptation. New adaptive sampling approaches and schemes are outlined. Illustrations suggest that these adaptive schemes can be used in real time with the potential for most efficient sampling.     

ERM learning with unbounded sampling

The learning approach of empirical risk minimization (ERM) is taken for the regression problem in the least square framework. A standard assumption for the error analysis in the literature is the uniform boundedness of the output sampling process. In this paper we abandon this boundedness assumption and conduct error analysis for the ERM learning algorithm with unbounded sampling processes satisfying an increment condition for the moments of the output. The key novelty of our analysis is a covering number argument for estimating the sample error.     

Learning and forgetting effects on a group scheduling problem

This paper proposes to investigate learning and forgetting effects on the problem of scheduling families of jobs on a single machine to minimize total completion time of jobs. A setup time is incurred whenever the single machine transfers job processing from a family to another family. To analyze the impact of learning and forgetting on this group scheduling problem, we structure three basic models and make some comparisons through computational experiments. The three models, including no forgetting, total forgetting and partial forgetting, assume that the processing time of a job is dependent on its position in a schedule. Some scheduling rules and a lower bound are derived in order to constitute our branch-and-bound algorithm for searching an optimal sequence. In addition, an efficient and simply-structured heuristic is also built to find a near-optimal schedule.     

以学生为主体的混合式个性化教学模式在分析化学课程中的探索与思考

In order to solve the issues of traditional teaching methods in the undergraduate basic theory course, the blended learning mode was put forward by taking analytical chemistry as an example. With concept of student-centered, personalized teaching, together with assistant means, was introduced to achieve the breakthrough in solving the problems. We hope to reinforce the adaptability and targeting of the course teaching.     

Analog circuits optimization based on evolutionary computation techniques

This paper presents a new design automation tool, based on a modified genetic algorithm kernel, in order to improve efficiency on the analog IC design cycle. The proposed approach combines a robust optimization with corner analysis, machine learning techniques and distributed processing capability able to deal with multi-objective and constrained optimization problems. The resulting optimization tool and the improvement in design productivity is demonstrated for the design of CMOS operational amplifiers.     

Shannon sampling II: Connections to learning theory

We continue our study [S. Smale, D.X. Zhou, Shannon sampling and function reconstruction from point values, Bull. Amer. Math. Soc. 41 (2004) 279–305] of Shannon sampling and function reconstruction. In this paper, the error analysis is improved. Then we show how our approach can be applied to learning theory: a functional analysis framework is presented; dimension independent probability estimates are given not only for the error in the L² spaces, but also for the error in the reproducing kernel Hilbert space where the learning algorithm is performed. Covering number arguments are replaced by estimates of integral operators.     

基于集成最小二乘支持向量机方法的近红外光谱分析

针对近红外（Near Infrared,NIR）光谱测量中的小样本问题。本文提出了一种集成最小二乘支持向量机（Ensemble Least Squares Support Vector Machine,ELS-SVM）新算法。首先使用随机子空间算法（Random Subspace Method,RSM）原始高维变量空间划分为若干个低维度的子空间,然后分别在各个子空间建立最小二乘支持向量机（LS-SVM）模型,最后构造一个集成结果来进行预测。针对一批柴油样本的实验结果表明,本法对柴油十六烷值的预测精度优于传统的LS-SVM方法。     

Development of an Instrument to Assess Attitudes Toward Science,Technology, Engineering,and Mathematics (STEM)

There is a need for more students to be interested in science, technology, engineering, and mathematics (STEM) careers to advance U.S. competitiveness and economic growth. A consensus exists that improving STEM education is necessary for motivating more students to pursue STEM careers. In this study, a survey to measure student (grades 4–6) attitudes toward STEM and STEM careers was developed and administered to 662 students from two STEM‐focused and three comprehensive (non‐STEM‐focused) schools. Cronbach's alphas for the whole survey and subscales indicated a high internal consistency. Statistically significant difference in means between students attending the STEM‐focused and comprehensive schools on the two subscales of the survey and the overall survey were found. However, the explained variance for these results was approximately 1%. The survey is a useful tool to assess efficacy of STEM education programs on student attitudes toward STEM and STEM careers.