Learning and Nonlinear Models最新文献_第10页

On Fast SVM Algorithms Used For Pattern Recognition 用于模式识别的快速SVM算法研究

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-vol4-no2-art1

Felipe A. C. de Bastos, M. Campos

{"title":"On Fast SVM Algorithms Used For Pattern Recognition","authors":"Felipe A. C. de Bastos, M. Campos","doi":"10.21528/LNLM-vol4-no2-art1","DOIUrl":"https://doi.org/10.21528/LNLM-vol4-no2-art1","url":null,"abstract":"This tutorial on fast Support Vector Machines (SVM) presents mathematical formulations and pseudocode Implementations of three algorithms used for fast SVM training. Traditional SVM training is a quadraticprogramming (QP) minimization problem that can be solved, e.g., using the Sequential Minimization Optimization (SMO) algorithm. This algorithm solves analytically a small QP optimization problem in each iteration, drastically reducing the training time needed by conventional QP optimizers. It is important to note that traditional SVM can be of two types: L1SVM and L2SVM, depending on the way that the training error is characterized in the SVM mathematical formulation. The SMO implementation presented in this tutorial applies only for the L1SVM, but it can be adapted to the L2SVM case. The Proximal SVM (PSVM) algorithm was also introduced as a fast alternative to traditional SVM classifiers that usually require a large amount of computation time for training. Unfortunately the PSVM algorithm may present poor performance due to biased optimal hyperplanes. The Unbiased Proximal SVM (UPSVM) algorithm uses a slightly different approach to circumvent this problem, such that an unbiased optimal hyperplane is always obtained. The results obtained show that the UPSVM algorithm performs better than the Sequential Minimal Optimization (SMO) algorithm with respect to training time with similar or better probability of correct pattern classification. The UPSVM algorithm also performs better than the PSVM algorithm with respect to probability of correct pattern classification (especially for low values of the regularization parameter C ), to training time, and to the number of floating point operations.","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132137728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SMART COMET: UMA ABORDAGEM AUTOMATIZADA PARA DETECÇÃO E CLASSIFICAÇÃO DE DANO CELULAR DISPONDO DO ENSAIO COMETA 智能彗星:一种基于彗星测试的细胞损伤检测和分类的自动化方法

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL14-NO1-ART3

K. W. Silva, B. Silva, Carlos Giovanni Nunes de Carvalho, F. M. M. Amaral

引用次数: 0

Predição da Variação Extrema do Nível do Mar Relacionada a Tempestades Severas Utilizando Redes Neurais Artificiais 利用人工神经网络预测与严重风暴相关的极端海平面变化

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL7-NO1-ART3

Marilia M. F. de Oliveira, N. Ebecken, Jorge Luiz Fernandes de Oliveira

{"title":"Predição da Variação Extrema do Nível do Mar Relacionada a Tempestades Severas Utilizando Redes Neurais Artificiais","authors":"Marilia M. F. de Oliveira, N. Ebecken, Jorge Luiz Fernandes de Oliveira","doi":"10.21528/LNLM-VOL7-NO1-ART3","DOIUrl":"https://doi.org/10.21528/LNLM-VOL7-NO1-ART3","url":null,"abstract":"This paper presents an Artificial Neural Network (ANN) model developed to predict extreme coastal sea level variation (storm surges) on Southeast Region of Brazil, related to the passage of frontal systems associated with extratropical cyclones that cause severe thunderstorms. Tidal forcing is the main cause of sea level daily variation but the effects of meteorological phenomenon are also present in rising and lowing of the observed sea level and tend to be more drastic accordingly to the event. Hourly time series of water level were used from two tide gauge station. 6-hourly series of atmospheric pressure and wind components from NCEP/NCAR reanalysis data set were also used on some grid points over the oceanic area. Correlations were verified to define the time lag between the meteorological variables and the coastal sea level response to the occurrences of the extreme atmospheric systems. These correlations and time lags were used as input variables of the ANN model. Simulations until 48 hours were tested with the neural model. This model was compared with multivariate linear regression and presented the best performance, generalizing the effect of the atmospheric interactions on extreme sea level variations.","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133583373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Indicação De Suspeitos De Irregularidade Em Instalações Elétricas De Baixa Tensão 低压电气装置可疑违规的迹象

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL6-NO1-ART2

Cyro Muniz, Karla Figueiredo, Marley M. B. R. Vellasco, M. A. C. Pacheco, Gustavo Chavez

引用次数: 2

ESTIMAÇÃO DE HARMÔNICAS NO CONTEXTO DA QUALIDADE DA ENERGIA ELÉTRICA UTILIZADA REDES NEURAIS ARTIFICIAIS 利用人工神经网络进行电能质量背景下的谐波估计

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-vol14-no1-art2

A. Pessoa, Eduardo Henrique de Oliveira Barbosa, P. Ulisses, H. M. G. C. Branco, R. Rabelo

引用次数: 1

Sistema de Trading no Mercado de Ações Implementado por Redes Neurais Artificiais 由人工神经网络实现的股票市场交易系统

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL11-NO2-ART3

Jarbas Aquiles Gambogi, O. L. Costa

引用次数: 1

Modelo Adaptativo Baseado Em Regras Nebulosas Aplicado À Previsão De Séries Temporais 基于模糊规则的自适应时间序列预测模型

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL5-NO1-ART4

I. Luna, S. Soares, R. Ballini

引用次数: 0

UMA ABORDAGEM PARA PROVISIONAMENTO AUTOMÁTICO DE SENSORES EM NUVENS DE SENSORES 传感器云中传感器自动配置的一种方法

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL14-NO1-ART1

M. Lemos, Carlos Giovanni Nunes de Carvalho, Douglas Lopes, Raimir Holanda, R. Rabelo

引用次数: 1

A Quantum-Inspired Evolutionary Learning Process to Design Dilation-Erosion Perceptrons for Financial Forecasting 基于量子启发的进化学习过程设计用于金融预测的膨胀-侵蚀感知器

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL10-NO3-ART6

Ricardo de A. Araújo, Adriano Oliveira, S. Soares, S. Meira

引用次数: 2

A Hybrid Method Using Evolutionary and a Linear Integer Model to Solve the Automatic Clustering Problem 基于进化和线性整数模型的自动聚类混合方法

Learning and Nonlinear Models Pub Date : 1900-01-01 DOI: 10.21528/LNLM-VOL13-NO2-ART2

Marcelo Dib Cruz, L. Ochi

引用次数: 1