某型无刷直流电动机的神经网络计算机控制设计

在无刷直流电机控制中,会受到负载等非线性因素的影响,实际应用时采用传统的控制策略往往获得的性能指标不理想.神经网络控制是智能控制的一个重要分支,为无刷直流电机的控制策略带来了新的思想和方法.论文设计了一种神经网络的学习算法及神经网络控制,该算法在某型无刷电机控制中显示了其优越性.     

一种基于分布式编码的同步梯度下降算法

基于数据并行化的异步随机梯度下降(ASGD)算法由于需要在分布式计算节点之间频繁交换梯度数据,从而影响算法执行效率.提出基于分布式编码的同步随机梯度下降(SSGD)算法,利用计算任务的冗余分发策略对每个节点的中间结果传输时间进行量化以减少单一批次训练时间,并通过数据传输编码策略的分组数据交换模式降低节点间的数据通信总量...     

基于非对称卷积-压缩激发-次代残差网络的人脸关键点检测

级联深度卷积神经网络（DCNN）算法为首先在人脸关键点检测中使用卷积神经网络（CNN）的模型,CNN的使用使得检测精度得到极大的提升。针对该策略需要对相邻阶段间的数据反复进行回归处理使得算法流程十分复杂的问题,提出基于非对称卷积-压缩激发-次代残差网络（AC-SE-ResNeXt）的人脸关键点检测算法。所提算法仅使用单阶段回归,既避免了级联策略中多阶段回归的算法流程复杂性,又解决了相邻阶段间数据需要进行预处理的问题。为了不降低精度,在次代残差网络（ResNeXt）块的基础上添加了非对称卷积（AC）模块和压缩激发（SE）模块,构建了AC-SE-ResNeXt网络模型。同时,为了能够精确拟合在不同光照、姿态、表情等复杂环境下的人脸,将AC-SE-ResNeXt网络模型加深到101层。对训练好的模型分别在数据集BioID和LFPW上进行测试,其中该模型在BioID数据集上的人脸五点关键点检测的综合平均误差率为1.99%,在LFPW数据集上的人脸五点关键点检测的综合平均误差率为2.3%。实验结果表明,所改进的算法不但简化了算法流程使之能进行端到端处理,而且其精度与级联DCNN算法相当,鲁棒性也有明显提升。     

基于Fuzzy ARTMAP神经网络的高分辨率图象土地覆盖分类及其评价

主要讨论了基于Fuzzy ARTMAP神经网络的高分辨率遥感图象土地覆盖分类方法及其实践．首先介绍了Fuzzy ARTMAP神经网络的原理，然后用SPOT XS图象试验数据进行土地覆盖分类．分类结果与传统的最大似然监督分类(MLC)、反馈式(Back Propagation，BP)神经网络的分类结果进行了比较．通过抽取500个样点对3种分类结果进行精度评价表明，Fuzzy ARTMAP神经网络相对其他两种方法，分类精度均有不同程度的改善，具有更好的分类结果，总分类精度比MLC和BP算法分别提高17．41％、7．32％．最后，对不同分类方法对于土地覆盖分类结果的影响进行了评价和分析．试验表明，Fuzzy ARTMAP神经网络用于高分辨图象土地覆盖分类研究可以获得相对较好的分类结果．     

基于预训练-微调策略的COVID-19预测模型

COVID-19的世界性大流行对整个社会产生了严重的影响,通过数学建模对确诊病例数进行预测将有助于为公共卫生决策提供依据。在复杂多变的外部环境下,基于深度学习的传染病预测模型成为研究热点。然而,现有模型对数据量要求较高,在进行监督学习时不能很好地适应低数据量的场景,导致预测精度降低。构建结合预训练-微调策略的COVID-19预测模型P-GRU。通过在源地区数据集上采用预训练策略,使模型提前获得更多的疫情数据,从而学习到COVID-19的隐式演变规律,为模型预测提供更充分的先验知识,同时使用包含最近历史信息的固定长度序列预测后续时间点的确诊病例数,并在预测过程中考虑本地人为限制政策因素对疫情趋势的影响,实现针对目标地区数据集的精准预测。实验结果表明,预训练策略能够有效提高预测性能,相比于卷积神经网络、循环神经网络、长短期记忆网络和门控循环单元模型,P-GRU模型在平均绝对百分比误差和均方根误差评价指标上表现优异,更适合用于预测COVID-19传播趋势。     

基于Levenberg-Marquardt算法的主机入侵检测系统研究

入侵检测系统是当前信息安全领域的研究热点,在保障信息安全方面起着重要的作用。对BP神经网络优化算法进行对比研究的基础上,利用Lvenberg-Marquardt 算法对传统BP算法进行改进,成功地将LMBP算法运用到基于Windows操作系统的主机入侵检测中去,建立LMBP-HIDS入侵检测系统模型。实验结果表明,运用Levenberg-Marquardt算法优化BP神经网络进行主机入侵检测．可以较好地提高学习速率,缩短训练过程。     

Breast Calcifications and Histopathological Analysis on Tumour Detection by CNN

The most salient argument that needs to be addressed universally is Early Breast Cancer Detection (EBCD), which helps people live longer lives. The Computer-Aided Detection (CADs)/Computer-Aided Diagnosis (CADx) system is indeed a software automation tool developed to assist the health professions in Breast Cancer Detection and Diagnosis (BCDD) and minimise mortality by the use of medical histopathological image classification in much less time. This paper purposes of examining the accuracy of the Convolutional Neural Network (CNN), which can be used to perceive breast malignancies for initial breast cancer detection to determine which strategy is efficient for the early identification of breast cell malignancies formation of masses and Breast microcalcifications on the mammogram. When we have insufficient data for a new domain that is desired to be handled by a pre-trained Convolutional Neural Network of Residual Network (ResNet50) for Breast Cancer Detection and Diagnosis, to obtain the Discriminative Localization, Convolutional Neural Network with Class Activation Map (CAM) has also been used to perform breast microcalcifications detection to find a specific class in the Histopathological image. The test results indicate that this method performed almost 225.15% better at determining the exact location of disease (Discriminative Localization) through breast microcalcifications images. ResNet50 seems to have the highest level of accuracy for images of Benign Tumour (BT)/Malignant Tumour (MT) cases at 97.11%. ResNet50’s average accuracy for pre-trained Convolutional Neural Network is 94.17%.     

基于神经网络离散混合蛙跳算法的多用户检测

为进一步提高基于离散混合蛙跳算法（DSFLA）的多用户检测性能,提出一种基于DSFLA和神经网络相结合的神经网络离散混合蛙跳算法,并用于多用户检测。在DSFLA的每一族内更新中,随机选择若干只“青蛙”采用Hopfield神经网络的寻优更新策略,进行快速迭代,寻找全局最优。仿真结果证明,基于神经网络离散混合蛙跳算法的多用户检测器在误码率、收敛速度、系统容量、抗远近能力等方面都优于传统方法和一些应用优化算法的多用户检测器。     

Resolution of pattern recognition problems using a hybrid Genetic/Random Neural Network learning algorithm

Gelenbe has proposed a neural network, called a Random Neural Network, which calculates the probability of activation of the neurons in the network. In this paper, we propose to solve the patterns recognition problem using a hybrid Genetic/Random Neural Network learning algorithm. The hybrid algorithm trains the Random Neural Network by integrating a genetic algorithm with the gradient descent rule-based learning algorithm of the Random Neural Network. This hybrid learning algorithm optimises the Random Neural Network on the basis of its topology and its weights distribution. We apply the hybrid Genetic/Random Neural Network learning algorithm to two pattern recognition problems. The first one recognises or categorises alphabetic characters, and the second recognises geometric figures. We show that this model can efficiently work as associative memory. We can recognise pattern arbitrary images with this algorithm, but the processing time increases rapidly.     

基于RBF神经网络的可疑交易监测模型

针对国内外金融领域可疑交易的低检测率问题,通过对RBF（Radial Basis Function）神经网络技术的分析与研究,提出了一种基于APC-III聚类算法和RLS（Recursive Least Square）算法的面向反洗钱的RBF神经网络模型并加以实现。APC-III聚类算法用于确定RBF神经网络隐含层的中心向量,RLS算法用来调整隐含层与输出层之间的连接权值。RBF神经网络与支持向量机（SVM）和孤立点检测相比,有更高的检测率和较低的误检率,因此,提出的模型具有重要的理论和实用价值。     

结合优先度排序神经网络的样本点分块拟合研究

样本点拟合是动态构造神经网络应用的重要研究领域。从输入输出映射的角度,依赖输出样本点优先关系,结合优先度排序神经网络把输入样本集合划分为不同子集,动态构造优先度排序神经网络对各个子样本集进行映射,对样本点进行分块并行神经网络构造,提高神经网络的训练速度。最后,通过对不同类型样本集进行测试,实验结果表明该算法能有效地减少拟合误差。     

在数据库中一种基于QPSO与BP神经网络的重复记录检测算法

在检测数据库重复记录的研究中,基于BP神经网络的检测（Duplicate Record Detection based on BP Neural Network,简称DRDBPNN）算法的性能与初始的参数设置有很大的关系,从而造成其性能不稳定的缺陷,因此本文提出了一种基于QPSO与BP神经网络的重复记录检测算法（Duplicate Record Detection based on Quantum Particle Swarm Optimization and BP Neural Network,简称DRDQPSQBPNN）。仿真表明,该算法能够有效地提升重复记录的检测效率。     

用基于视角的方法分析和合成图像

本文介绍了计算机视觉领域中基于视角的对象识别方法，对径向基函数(Radial Basis Function)神经网络模型进行了分析，分析了其训练算法，给出了一种利用径向基函数神经网络和基于视角的方法分析和合成图像的方法，并在MATLAB平台实现。仿真结果说明，采用RBF神经网络模型的训练速度和分析精度是令人满意的。     

Implementation of hybrid ANN–PSO algorithm on FPGA for harmonic estimation

Harmonic estimation is the main process in active filters for harmonic reduction. A hybrid Adaptive Neural Network–Particle Swarm Optimization (ANN–PSO) algorithm is being proposed for harmonic isolation. Originally Fourier Transformation is used to analyze a distorted wave. In order to improve the convergence rate and processing speed an Adaptive Neural Network Algorithm called Adaline has then been used. A further improvement has been provided to reduce the error and increase the fineness of harmonic isolation by combining PSO algorithm with Adaline algorithm. The inertia weight factor of PSO is combined along with the weight factor of Adaline and trained in Neural Network environment for better results. ANN–PSO provides uniform convergence with the convergence rate comparable that of Adaline algorithm. The proposed ANN–PSO algorithm is implemented on an FPGA. To validate the performance of ANN–PSO; results are compared with Adaline algorithm and presented herein.     

Hybrid intelligent systems for predicting software reliability

In this paper, we propose novel recurrent architectures for Genetic Programming (GP) and Group Method of Data Handling (GMDH) to predict software reliability. The effectiveness of the models is compared with that of well-known machine learning techniques viz. Multiple Linear Regression (MLR), Multivariate Adaptive Regression Splines (MARS), Backpropagation Neural Network (BPNN), Counter Propagation Neural Network (CPNN), Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS), TreeNet, GMDH and GP on three datasets taken from literature. Further, we extended our research by developing GP and GMDH based ensemble models to predict software reliability. In the ensemble models, we considered GP and GMDH as constituent models and chose GP, GMDH, BPNN and Average as arbitrators. The results obtained from our experiments indicate that the new recurrent architecture for GP and the ensemble based on GP outperformed all other techniques.     

一种基于动态权重的神经网络训练算法

提出一种基于动态粒子群算法的神经网络训练方法。神经网络权值选择是否合适直接关系到其非线性拟合能力,通过引入动态粒子群算法对神经网络进行训练,对神经网络各层连接权值进行优化。经过函数测试表明,相比粒子群算法,动态柱子群算法收敛速度更快且不易陷入局部,能更快更合理地训练神经网络从而优化网络连接权值。     

Forecasting financial time series volatility using Particle Swarm Optimization trained Quantile Regression Neural Network

Accurate forecasting of volatility from financial time series is paramount in financial decision making. This paper presents a novel, Particle Swarm Optimization (PSO)-trained Quantile Regression Neural Network namely PSOQRNN, to forecast volatility from financial time series. We compared the effectiveness of PSOQRNN with that of the traditional volatility forecasting models, i.e., Generalized Autoregressive Conditional Heteroskedasticity (GARCH) and three Artificial Neural Networks (ANNs) including Multi-Layer Perceptron (MLP), General Regression Neural Network (GRNN), Group Method of Data Handling (GMDH), Random Forest (RF) and two Quantile Regression (QR)-based hybrids including Quantile Regression Neural Network (QRNN) and Quantile Regression Random Forest (QRRF). The results indicate that the proposed PSOQRNN outperformed these models in terms of Mean Squared Error (MSE), on a majority of the eight financial time series including exchange rates of USD versus JPY, GBP, EUR and INR, Gold Price, Crude Oil Price, Standard and Poor 500 (S&P 500) Stock Index and NSE India Stock Index considered here. It was corroborated by the Diebold–Mariano test of statistical significance. It also performed well in terms of other important measures such as Directional Change Statistic (Dstat) and Theil's Inequality Coefficient. The superior performance of PSOQRNN can be attributed to the role played by PSO in obtaining the better solutions. Therefore, we conclude that the proposed PSOQRNN can be used as a viable alternative in forecasting volatility.     

Centroid Neural Network with Pairwise Constraints for Semi-supervised Learning

Neural Processing Letters - A clustering algorithm for datasets with pairwise constraints using the Centroid Neural Network (Cent.NN) is proposed in this paper. The proposed algorithm, referred to...     

广义神经网络系统与广义BP算法

本文在传统的神经网络理论基础上,将传统的神经元拓广为广义神经元,描述了以广义神经元为基础的广义神经网络系统的组成原理,提出了适应于广义神经网络系统的一种广义BP算法。并给出了该算法的数学推导,最后简要地介绍了广义神经网络系统原理在汉字识别中的应用。     

Real-Time Lossy Audio Signal Reconstruction Using Novel Sliding Based Multi-instance Linear Regression/Random Forest and Enhanced CGPANN

Neural Processing Letters - This paper proposes a novel NeuroEvolutionary algorithm called Enhanced Cartesian Genetic Programming evolved Artificial Neural Network (ECGPANN) as a predictor for the...