定量构效关系研究中分子描述符的相关性

分子描述符的计算是定量构效关系(QSAR)研究的基础,也是化学计量学研究的重要内容,在药物化学、环境科学等领域应用广泛。目前可计算的各种分子描述符已超过4 000种,如何选择合适的分子描述符,成为QSAR研究中首先要做的事。本文首先综述了分子描述符的研究进展,然后随机选择10 000余个化合物,利用Cerius2、Sybyl、Dragon软件计算1 600多种分子描述符,最后,采用相关分析和聚类分析研究各描述符之间的关系,相关分析表明,各类分子描述符之间存在较强的线性关系,这种关系通过聚类分析可以直观体现。     

区域高斯描绘子及其在物体识别中的应用

为了对几何变形的图像进行正确和有效的识别,对基于物体轮廓的高斯描绘子进行推广,构造了一种基于区域的新的不变量--区域高斯描绘子.构造思路主要有两点:①定义一个具有平移、尺度不变性的函数--区域高斯势函数;②分别计算区域高斯势函数在8个同心圆上的平均值,从而获得8个不变量.这些不变量不仅具有平移、旋转、尺度和反射不变性,而且对噪声不敏感,适用范围广.将区域高斯描绘子应用于物体识别,获得很高的识别率.     

基于压缩感知理论的3维人脸快速重建

形变模型是当前人脸重建研究中的一种主要方法。针对形变模型方法中模型构建的缺陷,提出一种基于压缩感知理论的快速三维人脸重建方法。首先,利用压缩感知理论估计三维原型人脸与目标人脸的形状相似性,根据相似性对原型样本进行筛选并构建相应的形变模型,提高建模精度和效率;然后,利用特征点信息进行稀疏模型匹配,并结合径向基函数插值重建生成特定的三维人脸,提高重建表面的平滑性。在BJUT三维数据库和CAS_PEAL二维数据库上的实验结果表明,与经典方法相比,本文方法能够有效地提高重建精度和速度,重建人脸具有较强真实感。     

一种结构自适应径向基函数神经网络

提出了一种新的结构自适应的径向基函数(RBF)神经网络模型。在该网络中,自组织映射(SOM)神经网络作为聚类网络,采用无监督学习算法对输入样本进行自组织分类,并将分类中心及其对应的权值向量传递给RBF神经网络,作为径向基函数的中心和相应的权值向量;RBF神经网络作为基础网络,采用高斯函数实现输入层到隐层的非线性映射,输出层则采用有监督学习算法训练网络的权值,从而实现输入层到输出层的非线性映射。通过对字母数据集进行仿真,表明该网络具有较好的性能。     

A self-organizing video neuro-tracker

We investigate a radial basis function neural network for the control of a video camera for tracking a target moving at high speed. The adjusting of the weights and parameters of the network minimizes the performance measureP that is the sum of the squared errors between the target position on the screen and the center of the screen over an entire trajectory of the target. Every adjustment of a single weight requires the evaluation of the performance measure over the fixed target trajectory. A very general target trajectory is required for generalized training of the network to appropriately control any camera and target situation. The results show considerable promise for this and similar cases where a trainable controller is desired. This work was presented, in part, at the International Symposium on Artificial Life and Robotics, Oita, Japan, February 18–20, 1996.     

基于径向基神经网络的浮游植物分类系统

应用模糊聚类和小波变换提取浮游植物活体的特征光谱,并以此为输入向量,引入径向基函数网络对浮游植物的光谱进行分类识别,建立了适用于光谱识别的径向基函数神经网络系统.结果表明,该方法较传统的统计方法更方便,识别准确率更高.     

基于改进自适应聚类算法的RBF神经网络分类器设计与实现

针对传统径向基函数神经网络构造的网络分类器通常存在分类精度不高、训练时间长等缺陷,首先提出了一种改进的自适应聚类算法,用于确定分类器的隐含层节点。该算法通过筛选基于轮廓系数的优秀样本群,来寻找最佳初始聚类中心,避免了传统K-means算法易受初始聚类中心点影响,导致最终的分类效果严重偏离全局等情况的发生。其次,将该改进算法用于构造径向基函数神经网络分类器和快速有效地确定隐含层节点径向基函数中心及函数的宽度。最后,通过大量UCI数据集的实验和仿真,验证了改进算法在聚类时间、聚类轮廓系数及聚类正确率等方面具有优越性。同时,大量的仿真实验也证明了基于改进算法构造的RBF分类器具有更高的分类精度。     

基于RBF神经网络的飞机发动机故障诊断研究

论述了径向基函数神经网络的基本网络结构和网络的学习及运行过程,结果表明:径向基神经网络具有极快的学习收敛速度。讨论了径向基神经网络在飞机发动机故障诊断中的应用,并对训练后的网络进行了仿真测试,仿真结果表明RBF网络有较高诊断正确率,且能满足实时诊断的要求。     

An efficient multi-objective optimization method for black-box functions using sequential approximate technique

Multi-objective optimization problems in practical engineering usually involve expensive black-box functions. How to reduce the number of function evaluations at a good approximation of Pareto frontier has been a crucial issue. To this aim, an efficient multi-objective optimization method based on a sequential approximate technique is suggested in this paper. In each iteration, according to the prediction of radial basis function with a micro multi-objective genetic algorithm, an extended trust region updating strategy is adopted to adjust the design region, a sample inheriting strategy is presented to reduce the number of new function evaluations, and then a local-densifying strategy is proposed to improve the accuracy of approximations in concerned regions. At the end of each iteration, the obtained actual Pareto optimal points are stored in an external archive and are updated as the iteration process. The effect of the present method is demonstrated by eight test functions. Finally, it is employed to perform the structure optimization of a vehicle door.     

基于混沌映射的非径向对称基函数的神经网络模型

为提高神经网络模型的预测精度,构建了非径向对称基函数神经网络模型结构。为确定非径向对称基函数神经网络模型参数,采用Ulam-von Neumann映射规则确定混沌变量,利用混沌变量的遍历性获得不同网络结构参数下的最优网络输出,以减少所构建网络模型的实际输出与期望输出的差值,并利用模型输出的误差变化率以决定是否增加新的隐层节点。给出基于混沌映射的非径向对称基函数的网络模型构建步骤。采用基于Mackey-Glass时滞微分方程的混沌时间序列预测问题验证该模型的预测精度,并同其他文献对该序列预测的精度以及所需隐层节点数作对比。比较结果表明,采用该设计模型具有对时间序列预测精度高且所需网络结构规模小等优点。     

一种优化的RBF神经网络模型用于网络流量预测

采用径向基RBF神经网络对网络流量数据的时间序列进行建模与预测。采用传统的学习算法对RBF网络训练时,对网络流量数据容易出现过拟合现象,提出了自适应量子粒子群优化AQPSO算法,用于训练RBF神经网络的基函数中心和宽度,并结合最小二乘法计算网络权值,改善了RBF神经网络的泛化能力。实验结果表明,采用AQPSO算法获得的RBF神经网络模型具有泛化能力强、稳定性良好的特点,在网络流量预测中有一定的实用价值。     

组合网络式距离保护

传统距离保护受到不同故障条件的影响,而且,有时会产生继电器的误动作.本文提出一种组合网络式距离保护,该方法将BP网络与径向基函数网络的优点相结合.仿真结果表明,该方法能够对电力系统的各种故障情况进行更正确、迅速的识别.     

Modelling of a magneto-rheological damper by evolving radial basis function networks

This paper presents an approach to approximate the forward and inverse dynamic behaviours of a magneto-rheological (MR) damper using evolving radial basis function (RBF) networks. Due to the highly nonlinear characteristics of MR dampers, modelling of MR dampers becomes a very important problem to their applications. In this paper, an alternative representation of the MR damper in terms of evolving RBF networks, which have a structure of four input neurons and one output neuron to emulate the forward and inverse dynamic behaviours of an MR damper, respectively, is developed by combining the genetic algorithms (GAs) to search for the network centres with other standard learning algorithms. Training and validating of the evolving RBF network models are achieved by using the data generated from the numerical simulation of the nonlinear differential equations proposed for the MR damper. It is shown by the validation tests that the evolving RBF networks can represent both forward and inverse dynamic behaviours of the MR damper satisfactorily.     

一种基于高斯核的RBF神经网络学习算法

RBF神经网络中心等参数确定得是否合理将直接影响到RBF神经网络的学习性能。通过有监督学习的方法来确定RBF神经网络的中心等参数是最一般化的方法。在这种方法中,参数的初始化问题是关键问题。文章在分析RBF神经网络映射性能的基础上,提出了中心等参数初始化的一种方法,并借助于梯度下降法给出了RBF神经网络的学习算法。多种实例表明,所给出的学习算法是有效的。该研究为RBF神经网络的广泛应用提供了一定的技术保障。     

Interactive 3D medical image segmentation with energy-minimizing implicit functions

We present an interactive segmentation method for 3D medical images that reconstructs the surface of an object using energy-minimizing, smooth, implicit functions. This reconstruction problem is called variational interpolation. For an intuitive segmentation of medical images, variational interpolation can be based on a set of user-drawn, planar contours that can be arbitrarily oriented in 3D space. This also allows an easy integration of the algorithm into the common manual segmentation workflow, where objects are segmented by drawing contours around them on each slice of a 3D image.Because variational interpolation is computationally expensive, we show how to speed up the algorithm to achieve almost real-time calculation times while preserving the overall segmentation quality. Moreover, we show how to improve the robustness of the algorithm by transforming it from an interpolation to an approximation problem and we discuss a local interpolation scheme.A first evaluation of our algorithm by two experienced radiology technicians on 15 liver metastases and 1 liver has shown that the segmentation times can be reduced by a factor of about 2 compared to a slice-wise manual segmentation and only about one fourth of the contours are necessary compared to the number of contours necessary for a manual segmentation.     

Learning experiments with genetic optimization of a generalized regression neural network

This paper reports a study unifying optimization by genetic algorithm with a generalized regression neural network. Experiments compare hill-climbing optimization with that of a genetic algorithm, both in conjunction with a generalized regression neural network. Controlled data with nine independent variables are used in combination with conjunctive and compensatory decision forms, having zero percent and 10 percent noise levels. Results consistently favor the GRNN unified with the genetic algorithm.     

Adaptive statistical algorithms in network reliability analysis

The paper is concerned with introducing novel algorithms, such as adaptive approximation and deterministic radial basis function (RBF) method, for calculating the average loss (AL). Different approximators are trained to approximate the loss function and, after a short learning period, AL can be evaluated analytically with fast calculations. An improvement of the Li–Silvester (LS) method is also presented which yields a sharper lower bound on AL. The efficiency of the new methods are proven by theoretical analysis as well as demonstrated by excessive simulations.