AABC算法优化ELM的心脏病辅助诊断

针对ELM算法在心脏病辅助诊断中分类精度不高的缺陷,提出自适应人工蜂群算法优化ELM隐层输入权值和偏置的心脏病辅助诊断方法。采用自适应遗传算法对数据进行特征选择,以最优特征子集构造样本输入自适应人工蜂群算法优化ELM的分类模型。自适应人工蜂群算法改进原算法的跟随蜂概率选择机制,在搜索阶段引入最优解与次优解,通过自适应算子调整二者的引导作用。仿真结果表明,该方法相比于其它方法提高了分类精度,减少了总体耗时。     

磷酸铁锂电池的SOC预测

电池荷电状态(SOC)准确预测是电池管理系统的关键任务.针对过去电池SOC预测精度低等问题,提出了一种采用极限学习机神经网络(ELM)的预测模型,以电池电压和电流作为模型的输入量,SOC作为输出量.在建模过程中,采用粒子群优化算法(PSO)对ELM随机给定的输入权值矩阵和隐层阈值进行寻优,降低了随机性给模型造成的影响,提高了模型预测精度.利用实验采集的数据进行模型训练和预测,结果表明,用粒子群算法优化后的极限学习机模型(PSOELM)与单纯的ELM以及传统的BP和SVM相比,具有更高的预测精度和泛化性能.为磷酸铁锂电池的SOC预测提供了一种新的方法.     

基于ELM和FOA的股票价格预测

针对股票价格预测中应用极限学习机预测存在稳定性不理想的问题,提出了一种改进果蝇优化极限学习机（IFOA-ELM）预测模型的算法。在该算法中,果蝇群通过不断调整群半径来优化ELM的输入层与隐含层连接权值和隐含层阈值,并以优化后的结果为基础,构建ELM预测模型。将IFOA-ELM模型用于股票价格预测。实验表明,与ELM和FOA-ELM相比,IFOA-ELM在股票价格预测中具有更高的预测精度和更好的稳定性。     

基于训练图CNN特征的视频人体动作识别算法

为将卷积神经网络(CNN)应用到视频理解中,提出一种基于训练图CNN特征的识别算法。利用图像RGB数据识别视频人体动作,使用现有的CNN模型从图像中提取特征,并采用长短记忆单元的递归神经网络进行训练分类,研究CNN模型和隐层的选择、优化、特征矢量化和降维。实验结果表明,与使用图像RGB数据注意力模型的算法和组合长短期记忆模型算法相比,该算法具有更高的准确率。     

基于结合混沌纵横交叉的粒子群算法优化极限学习机的短期负荷预测

为了解决传统的单一负荷预测模型精度低以及常规智能算法在解决高维、多模复杂问题时容易陷入局部最优的问题,提出了一种结合混沌纵横交叉的粒子群算法（CC-PSO）优化极限学习机(ELM)的短期负荷预测模型。ELM的泛化能力与其输入权值和隐含层偏置密切相关,采用结合混沌纵横交叉的粒子群算法优化ELM的输入权值与隐含层偏置,提高了ELM的泛化能力和预测精度。选择广东某地区实际电网负荷数据进行分析,研究结果表明,相对于BP神经网络和支持向量机,ELM具有更高的泛化能力和预测精度;CC-PSO相对于粒子群和遗传算法具有更高的全局搜索能力,CC-PSO-ELM模型具有较高的负荷预测精度。     

用于大数据分类的快速隐层优化分布式极限学习机

针对大数据分类问题应用设计了一种快速隐层优化方法来解决分布式超限学习机（Extreme Learning Machine,ELM）在训练过程中存在的突出问题--需要独立重复运行多次才能优化隐层结点个数或模型泛化性能。在不增加算法时间复杂度的前提下,新算法能同时训练多个ELM隐层网络,全面兼顾模型泛化能力和隐层结点个数的优化,并通过分布式计算避免大量重复计算。同时,在算法求解过程中通过这种方式能更精确、更直观地学习隐含层结点个数变化带来的影响。比较多种类型标准测试函数的实验结果,相对于分布式ELM,新算法在求解精度、泛化能力、稳定性上大大提高。     

基于蚁群算法优化回声状态网络的研究

针对输出权值采用最小二乘法的回声状态网络(ESN),在随机选取输入权值和隐层神经元阈值时,存在收敛速度慢、预测精度不稳定等问题,提出了基于蚁群算法优化回声状态网络(ACO-ESN)的算法。该算法将优化回声状态网络的初始输入权值、隐层神经元阈值问题转化为蚁群算法中蚂蚁寻找最佳路径的问题,输出权值采用最小二乘法计算,通过蚁群算法的更新、变异、遗传等操作训练回声状态网络,选择出使回声状态网络预测误差最小的输入权值和阈值,从而提高其预测性能。将ACO-ESN与ELM、I-ELM、OS-ELM、B-ELM等神经网络的仿真结果进行对比,结果验证经过蚁群算法优化的回声状态网络加快了其收敛速度,改善了其预测性能,并增强了隐层神经元的敏感度。     

一种粒子群优化的SVM-ELM模型

极限学习机(extreme learning machine,ELM)是一种简单易用、有效的单隐层前馈神经网络(single hidden layer feedforward neural networks,SLFNs)学习算法,近几年来已成为机器学习研究的热门领域之一。但是ELM单个隐层节点的判断能力不足,分类正确率的高低在一定程度上取决于隐层节点数。为了提高ELM单个隐层节点的判断能力,将支持向量机(support vector machine,SVM)和ELM结合,建立一种精简的SVM-ELM模型。同时,该模型为了避免人为选择参数的主观性,利用粒子群算法(particle swarm optimization,PSO)的全局搜索最优解对参数进行自动优化选取,建立了PSO-SVM-ELM模型。实验证明,该模型较SVMELM和ELM分类精度有较大的提高,具有很好的稳健性和泛化性。     

极限学习机的分类器集成模型研究

将极限学习机算法与旋转森林算法相结合,提出了以ELM算法为基分类器并以旋转森林算法为框架的RF-ELM集成学习模型。在8个数据集上进行了3组预测实验,根据实验结果讨论了ELM算法中隐含层神经元个数对预测结果的影响以及单个ELM模型预测结果不稳定的缺陷;将RF-ELM模型与单ELM模型和基于Bagging算法集成的ELM模型相比较,由稳定性和预测精度的两组对比实验的实验结果表明,对ELM的集成学习可以有效地提高ELM模型的性能,且RF-ELM模型较其他两个模型具有更好的稳定性和更高的准确率,验证了RF-ELM是一种有效的ELM集成学习模型。     

基于影响度剪枝的极端学习机分类算法研究

针对极端学习机(ELM)网络规模控制问题,从剪枝思路出发,提出了一种基于影响度剪枝的ELM分类算法。利用ELM网络单个隐节点连接输入层和输出层的权值向量、该隐节点的输出、初始隐节点个数以及训练样本个数,定义单个隐节点相对于整个网络学习的影响度,根据影响度判断隐节点的重要性并将其排序,采用与ELM网络规模相匹配的剪枝步长删除冗余节点,最后更新隐含层与输入层和输出层连接的权值向量。通过对多个UCI机器学习数据集进行分类实验,并将提出的算法与EM-ELM、PELM和ELM算法相比较,结果表明,该算法具有较高的稳定性和测试精度,训练速度较快,并能有效地控制网络规模。     

基于局部保留投影的堆叠隐空间模糊C均值算法*

传统模糊聚类算法在处理复杂非线性数据时学习能力较差。针对此问题,文中基于极限学习机(ELM)理论,结合局部保留投影(LPP)与ELM特征映射,提出压缩隐空间特征映射算法,从而将原始数据从原空间映射至压缩ELM隐空间中。通过连接多个压缩隐空间特征映射,结合模糊聚类技术,提出基于LPP的堆叠隐空间模糊C均值算法。大量实验表明,文中算法对模糊指数的变化不敏感,在处理复杂非线性数据和存在类内差异的图像数据时,能够取得更精确、高效、稳定的学习效果。     

An optimized extreme learning machine-based novel model for bearing fault classification

This work addresses the rolling element bearing (REB) fault classification problem by tackling the issue of identifying the appropriate parameters for the extreme learning machine (ELM) and enhancing its effectiveness. This study introduces a memetic algorithm (MA) to identify the optimal ELM parameter set for compact ELM architecture alongside better ELM performance. The goal of using MA is to investigate the promising solution space and systematically exploit the facts in the viable solution space. In the proposed method, the local search method is proposed along with link-based and node-based genetic operators to provide a tight ELM structure. A vibration data set simulated from the bearing of rotating machinery has been used to assess the performance of the optimized ELM with the REB fault categorization problem. The complexity involved in choosing a promising feature set is eliminated because the vibration data has been transformed into kurtograms to reflect the input of the model. The experimental results demonstrate that MA efficiently optimizes the ELM to improve the fault classification accuracy by around 99.0% and reduces the requirement of hidden nodes by 17.0% for both data sets. As a result, the proposed scheme is demonstrated to be a practically acceptable and well-organized solution that offers a compact ELM architecture in comparison to the state-of-the-art methods for the fault classification problem.     

堆叠隐空间模糊C 均值聚类算法

基于极限学习机理论, 将主成分分析技术与ELM特征映射相结合, 提出一种基于主成分分析的压缩隐空间构建新方法. 结合多层神经网络学习方法对隐空间进行多层融合, 进一步提出了堆叠隐空间模糊C 均值聚类算法,从而提高对非线性数据的学习能力. 实验结果表明, 所提出算法在处理复杂非线性数据时更加高效、稳定, 同时克服了模糊聚类算法对模糊指数的敏感性问题.

     

基于ELM的局部空间信息的模糊C均值聚类图像分割算法

极限学习机（Extreme learning machine, ELM）作为一种新技术具有在回归和分类中良好的泛化性能。局部空间信息的模糊C均值算法(Weighted fuzzy local information C-means, WFLICM)用邻域像素点的空间信息标记中心点的影响因子,增强了模糊C均值聚类算法的去噪声能力。基于极限学习机理论,对WFLICM进行改进优化,提出了基于ELM的局部空间信息的模糊C均值聚类图像分割算法(New kernel weighted fuzzy local information C-means based on ELM,ELM-NKWFLICM)。该方法基于ELM特征映射技术,将原始数据通过ELM特征映射技术映射到高维ELM隐空间中,再用改进的新核局部空间信息的模糊C均值聚类图像分割算法(New kernel weighted fuzzy local information C-means,NKWFLICM)进行聚类。 实验结果表明 ELM-NKWFLICM算法具有比WFLICM算法更强的去噪声能力,且很好地保留了原图像的细节,算法在处理复杂非线性数据时更高效, 同时克服了模糊聚类算法对模糊指数的敏感性问题。     

Uncertainty evaluation and model selection of extreme learning machine based on Riemannian metric

Considering the uncertainty of hidden neurons, choosing significant hidden nodes, called as model selection, has played an important role in the applications of extreme learning machines(ELMs). How to define and measure this uncertainty is a key issue of model selection for ELM. From the information geometry point of view, this paper presents a new model selection method of ELM for regression problems based on Riemannian metric. First, this paper proves theoretically that the uncertainty can be characterized by a form of Riemannian metric. As a result, a new uncertainty evaluation of ELM is proposed through averaging the Riemannian metric of all hidden neurons. Finally, the hidden nodes are added to the network one by one, and at each step, a multi-objective optimization algorithm is used to select optimal input weights by minimizing this uncertainty evaluation and the norm of output weight simultaneously in order to obtain better generalization performance. Experiments on five UCI regression data sets and cylindrical shell vibration data set are conducted, demonstrating that the proposed method can generally obtain lower generalization error than the original ELM, evolutionary ELM, ELM with model selection, and multi-dimensional support vector machine. Moreover, the proposed algorithm generally needs less hidden neurons and computational time than the traditional approaches, which is very favorable in engineering applications.     

基于极限学习机的股票价格预测

极限学习机（ Extreme Learning Machine , ELM）是一种新型的单馈层神经网络算法,克服了传统的误差反向传播方法需要多次迭代,算法的计算量和搜索空间大的缺点,只需要设置合适的隐含层节点个数,为输入权和隐含层偏差进行随机赋值,一次完成无需迭代。研究表明股票市场是一个非常复杂的非线性系统,需要用到人工智能理论、统计学理论和经济学理论。本文将极限学习机方法引入股票价格预测中,通过对比支持向量机（ Support Vector Machine , SVM）和误差反传神经网络（ Back Propagation Neural Network , BP神经网络）,分析极限学习机在股票价格预测中的可行性和优势。结果表明极限学习机预测精度高,并且在参数选择及训练速度上具有较明显的优势。     

Dynamic action recognition based on dynemes and Extreme Learning Machine

In this paper, we propose a novel method that performs dynamic action classification by exploiting the effectiveness of the Extreme Learning Machine (ELM) algorithm for single hidden layer feedforward neural networks training. It involves data grouping and ELM based data projection in multiple levels. Given a test action instance, a neural network is trained by using labeled action instances forming the groups that reside to the test sample’s neighborhood. The action instances involved in this procedure are, subsequently, mapped to a new feature space, determined by the trained network outputs. This procedure is performed multiple times, which are determined by the test action instance at hand, until only a single class is retained. Experimental results denote the effectiveness of the dynamic classification approach, compared to the static one, as well as the effectiveness of the ELM in the proposed dynamic classification setting.     

Performance enhancement of extreme learning machine for multi-category sparse data classification problems

This paper presents a performance enhancement scheme for the recently developed extreme learning machine (ELM) for multi-category sparse data classification problems. ELM is a single hidden layer neural network with good generalization capabilities and extremely fast learning capacity. In ELM, the input weights are randomly chosen and the output weights are analytically calculated. The generalization performance of the ELM algorithm for sparse data classification problem depends critically on three free parameters. They are, the number of hidden neurons, the input weights and the bias values which need to be optimally chosen. Selection of these parameters for the best performance of ELM involves a complex optimization problem.In this paper, we present a new, real-coded genetic algorithm approach called ‘RCGA-ELM’ to select the optimal number of hidden neurons, input weights and bias values which results in better performance. Two new genetic operators called ‘network based operator’ and ‘weight based operator’ are proposed to find a compact network with higher generalization performance. We also present an alternate and less computationally intensive approach called ‘sparse-ELM’. Sparse-ELM searches for the best parameters of ELM using K-fold validation. A multi-class human cancer classification problem using micro-array gene expression data (which is sparse), is used for evaluating the performance of the two schemes. Results indicate that the proposed RCGA-ELM and sparse-ELM significantly improve ELM performance for sparse multi-category classification problems.     

改进人工蜂群优化的极限学习机

针对极限学习机(ELM)在训练过程中需要大量隐含层节点的问题,提出了差分进化与克隆算法改进人工蜂群优化的极限学习机(DECABC-ELM),在人工蜂群算法的基础上,引入了差分进化算法的差分变异算子和免疫克隆算法的克隆扩增算子,改进了人工蜂群收敛速度慢等缺点,使用改进的人工蜂群算法计算ELM的隐含层节点参数.将算法应用于回归和分类数据集,并与其他算法进行比较,获得了良好的效果.     

Performance enhancement of extreme learning machine for power system disturbances classification

This paper proposes an optimal feature and parameter selection approach for extreme learning machine (ELM) for classifying power system disturbances. The relevant features of non-stationary time series data from power disturbances are extracted using a multiresolution S-transform which can be treated either as a phase corrected wavelet transform or a variable window short-time Fourier transform. After extracting the relevant features from the time series data, an integrated PSO and ELM architectures are used for pattern recognition of disturbance waveform data. The particle swarm optimization is a powerful meta-heuristic technique in artificial intelligence field; therefore, this study proposes a PSO-based approach, to specify the beneficial features and the optimal parameter to enhance the performance of ELM. One of the advantages of ELM over other methods is that the parameter that the user must properly adjust is the number of hidden nodes only. In this paper, a hybrid optimization mechanism is proposed which combines the discrete-valued PSO with the continuous-valued PSO to optimize the input feature subset selection and the number of hidden nodes to enhance the performance of ELM. The experimental results showed the proposed algorithm is faster and more accurate in discriminating power system disturbances.