Locally Weighted Learning

This paper surveys locally weighted learning, a form of lazy learning and memory-based learning, and focuses on locally weighted linear regression. The survey discusses distance functions, smoothing parameters, weighting functions, local model structures, regularization of the estimates and bias, assessing predictions, handling noisy data and outliers, improving the quality of predictions by tuning fit parameters, interference between old and new data, implementing locally weighted learning efficiently, and applications of locally weighted learning. A companion paper surveys how locally weighted learning can be used in robot learning and control.     

线性时变系统的区间稳定性与鲁棒稳定性

本文应用向量比较定理研究线性时变系统的区间稳定性和具非线性时变摄动的线性时变系统的鲁棒稳定性，所得的新结果包含文献的一些主要结果作为特例，本文的研究方法说明向量比较方法是分析区间稳定性和鲁棒稳定性的一种自然而有力的工具。     

f—x域内提高地震资料的信噪比

在 f-x 域,地震记录可以表示成有限个调频调幅正弦波的叠合,叠合的结果是一个连续且光滑的函数。我们可以用一个简单的 n 次多项式去逼近这个函数。本文提出了一种在 f-x 域用最小平方拟合原理消除随机噪声的新方法,并证明了在噪声为自噪以及空间采样足够密的条件下,拟合结果与噪声无关,即 f-x 域最小平方拟合处理可以有效地消除记录上的随机白噪,而对有效信号的破坏甚微。由于理论上允许振幅值及同相轴形态在空间上发生变化,因而该方法不具有混波作用,是高分辨率处理的理想工具。     

锚网支护在综采工作面开切眼中的应用

介绍了平煤集团一矿在戊8— 310 40综合机械化采煤工作面开切眼实施煤巷锚网支护的基本情况。总结了经验 ,为同类条件下的锚网支护设计提供了依据。     

Practical method for generation expansion planning based on dynamic programming

This paper presents an efficient method which provides the optimal generation mix and the optimal generation construction process. The approximation method in which the dynamic programming technique and gradient method are combined is applied to determine the optimal generation mix with hydropower generation technologies. The successive approximations dynamic programming (SADP) technique, which is very suitable for high-dimensional multistage decision process problems, is used for obtaining the optimal generation construction process. The effectiveness and feasibility of the developed technique are demonstrated on a practical power system model which has five types of generation technologies including a hydropower generation technology.     

钢带捆扎对木箱强度的影响

从力学角度分析了钢带捆扎对木箱强度的影响，指出木箱经钢带捆扎后，可减小箱底的最大弯矩，提高木箱的承载能力和减少板厚。同时还在无支撑间隔的基础上，详细介绍了钢带捆扎的设计方法。     

声表面波传感器电路参数的CAD方法

本文提出了声表面波传感器电路参数的一种ＣＡＤ方法，即有约束内点惩罚函数－鲍威尔－阻尼最小二乘法，并对两个实例进行了计算。     

Melt index prediction by weighted least squares support vector machines

Melt index is considered an important quality variable determining product specifications. Reliable prediction of melt index (MI) is crucial in quality control of practical propylene polymerization processes. In this paper a least squares support vector machines (LS‐SVM) soft‐sensor model of propylene polymerization process is developed to infer the MI of polypropylene from other process variables. Considering the use of a SSE cost function without regularization might lead to less robust estimates; the weighted least squares support vector machines (weighted LS‐SVM) approach of propylene polymerization process is further proposed to obtain a robust estimation of melt index. The performance of standard SVM model is taken as a basis of comparison. A detailed comparison research among the standard SVM, LS‐SVM, and weighted LS‐SVM models is carried out. The research results confirm the effectiveness of the presented methods. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 101: 285–289, 2006     

Effect of carbon black composite (CBC) support properties on hydrodesulfurization performance of sulfided Mo and Co, and carburized Mo, catalysts

Carbon black composites (CBCs) have been prepared by pyrolyzing mixture of a carbon black with polyfurfuryl alcohol and then pretreated by oxidation with nitric acid, gasification with water steam or ammoxidation. The effects of the chemical character of the carrier surface, nature of the active metal phase and pH value of the impregnation solution on the catalytic activity towards the hydrodesulfurization (HDS) of thiophene of the CBC supported Mo (Co) catalysts were determined. It was stated that the catalytic properties of the CBC supported sulfides of Mo or Co and of Mo carbides are affected by the chemical character of the carrier surface. Generally, catalysts supported over basic surface CBC exhibit higher activity than those ones supported over CBC possessing acidic surface character. Co catalysts supported on acidic surface show lower activity (per mol of active metal) than Mo based ones supported on the same carrier. In the case of catalysts supported on basic CBC, Co exhibits distinctly higher activity than Mo. At the experimental conditions adopted for this study, CBC surface properties, active phase nature, and catalyst impregnation pH were found to exert a relatively small influence on both HDS and hydrogenation activities.     

An efficient clustering scheme using support vector methods

Support vector clustering involves three steps—solving an optimization problem, identification of clusters and tuning of hyper-parameters. In this paper, we introduce a pre-processing step that eliminates data points from the training data that are not crucial for clustering. Pre-processing is efficiently implemented using the R*-tree data structure. Experiments on real-world and synthetic datasets show that pre-processing drastically decreases the run-time of the clustering algorithm. Also, in many cases reduction in the number of support vectors is achieved. Further, we suggest an improvement for the step of identification of clusters.