Learning and Nonlinear Models最新文献_第4页

Using the Kullback-Leibler Divergence and Kolmogorov-Smirnov Test to Select Input Sizes to the Fault Diagnosis Problem Based on a CNN Model 利用Kullback-Leibler散度和Kolmogorov-Smirnov检验选择基于CNN模型的故障诊断问题的输入大小

Learning and Nonlinear Models Pub Date : 2021-06-30 DOI: 10.21528/LNLM-VOL18-NO2-ART2

R. Monteiro, C. Bastos-Filho, M. Cerrada, Diego Cabrera, Réne-Vinicio Sánchez

{"title":"Using the Kullback-Leibler Divergence and Kolmogorov-Smirnov Test to Select Input Sizes to the Fault Diagnosis Problem Based on a CNN Model","authors":"R. Monteiro, C. Bastos-Filho, M. Cerrada, Diego Cabrera, Réne-Vinicio Sánchez","doi":"10.21528/LNLM-VOL18-NO2-ART2","DOIUrl":"https://doi.org/10.21528/LNLM-VOL18-NO2-ART2","url":null,"abstract":"Choosing a suitable size for signal representations, e.g., frequency spectra, in a given machine learning problem is not a trivial task. It may strongly affect the performance of the trained models. Many solutions have been proposed to solve this problem. Most of them rely on designing an optimized input or selecting the most suitable input according to an exhaustive search. In this work, we used the Kullback-Leibler Divergence and the Kolmogorov-Smirnov Test to measure the dissimilarity among signal representations belonging to equal and different classes, i.e., we measured the intraclass and interclass dissimilarities. Moreover, we analyzed how this information relates to the classifier performance. The results suggested that both the interclass and intraclass dissimilarities were related to the model accuracy since they indicate how easy a model can learn discriminative information from the input data. The highest ratios between the average interclass and intraclass dissimilarities were related to the most accurate classifiers. We can use this information to select a suitable input size to train the classification model. The approach was tested on two data sets related to the fault diagnosis of reciprocating compressors.","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"1044 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113995608","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

Evolving deep neural networks for Time Series Forecasting 用于时间序列预测的进化深度神经网络

Learning and Nonlinear Models Pub Date : 2021-06-30 DOI: 10.21528/LNLM-VOL18-NO2-ART4

Lídio Mauro Lima de Campos, J. H. A. Pereira, Danilo Souza Duarte, R. C. L. Oliveira

引用次数: 1

A Study of the Influence of Data Complexity and Similarity on Soft Biometrics Classification Performance in a Transfer Learning Scenario 迁移学习场景下数据复杂度和相似度对软生物特征分类性能的影响研究

Learning and Nonlinear Models Pub Date : 2021-06-30 DOI: 10.21528/LNLM-VOL18-NO2-ART5

M. Romero, M. Gutoski, L. T. Hattori, Manassés Ribeiro, H. S. Lopes

{"title":"A Study of the Influence of Data Complexity and Similarity on Soft Biometrics Classification Performance in a Transfer Learning Scenario","authors":"M. Romero, M. Gutoski, L. T. Hattori, Manassés Ribeiro, H. S. Lopes","doi":"10.21528/LNLM-VOL18-NO2-ART5","DOIUrl":"https://doi.org/10.21528/LNLM-VOL18-NO2-ART5","url":null,"abstract":"Transfer learning is a paradigm that consists in training and testing classifiers with datasets drawn from distinct distributions. This technique allows to solve a particular problem using a model that was trained for another purpose. In the recent years, this practice has become very popular due to the increase of public available pre-trained models that can be fine-tuned to be applied in different scenarios. However, the relationship between the datasets used for training the model and the test data is usually not addressed, specially where the fine-tuning process is done only for the fully connected layers of a Convolutional Neural Network with pre-trained weights. This work presents a study regarding the relationship between the datasets used in a transfer learning process in terms of the performance achieved by models complexities and similarities. For this purpose, we fine-tune the final layer of Convolutional Neural Networks with pre-trained weights using diverse soft biometrics datasets. An evaluation of the performances of the models, when tested with datasets that are different from the one used for training the model, is presented. Complexity and similarity metrics are also used to perform the evaluation.","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134299758","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: 3

Comparative Analysis of Convolutional Neural Networks Applied in the Detection of Pneumonia Through X-Ray Images of Children 卷积神经网络在儿童x线图像肺炎检测中的应用对比分析

Learning and Nonlinear Models Pub Date : 2021-06-30 DOI: 10.21528/LNLM-VOL18-NO2-ART1

L. Silva, L. Araújo, Victor F. Souza, Raimundo Matos Barros Neto, Adam Santos

{"title":"Comparative Analysis of Convolutional Neural Networks Applied in the Detection of Pneumonia Through X-Ray Images of Children","authors":"L. Silva, L. Araújo, Victor F. Souza, Raimundo Matos Barros Neto, Adam Santos","doi":"10.21528/LNLM-VOL18-NO2-ART1","DOIUrl":"https://doi.org/10.21528/LNLM-VOL18-NO2-ART1","url":null,"abstract":"Pneumonia is one of the most common medical problems in clinical practice and is the leading fatal infectious disease worldwide. According to the World Health Organization, pneumonia kills about 2 million children under the age of 5 and is constantly estimated to be the leading cause of infant mortality, killing more children than AIDS, malaria, and measles combined. A key element in the diagnosis is radiographic data, as chest x-rays are routinely obtained as a standard of care and can aid to differentiate the types of pneumonia. However, a rapid radiological interpretation of images is not always available, particularly in places with few resources, where childhood pneumonia has the highest incidence and mortality rates. As an alternative, the application of deep learning techniques for the classification of medical images has grown considerably in recent years. This study presents five implementations of convolutional neural networks (CNNs): ResNet50, VGG-16, InceptionV3, InceptionResNetV2, and ResNeXt50. To support the diagnosis of the disease, these CNNs were applied to solve the classification problem of medical radiographs from people with pneumonia. InceptionResNetV2 obtained the best recall and precision results for the Normal and Pneumonia classes, 93.95% and 97.52% respectively. ResNeXt50 achieved the best precision and f1-score results for the Normal class (94.62% and 94.25% respectively) and the recall and f1-score results for the Pneumonia class (97.80% and 97.65%, respectively).","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122189230","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: 2

Methodology for Classifying Diseases in Plants Using Convolutional Neural Networks 基于卷积神经网络的植物病害分类方法

Learning and Nonlinear Models Pub Date : 2021-06-30 DOI: 10.21528/LNLM-VOL18-NO2-ART3

Michel Costa, Vanessa Castro Rezende, Cledisson Martins, Adam Santos

引用次数: 0

Improving Prototypes Representativeness by Internal Validity Index Analysis 利用内部效度指标分析提高原型的代表性

Learning and Nonlinear Models Pub Date : 2021-01-01 DOI: 10.21528/lnlm-vol19-no1-art1

Alexandre Szabo, Thomaz A. Ruckl

引用次数: 0

Carteiras de Black Litterman com Análises Baseadas em Redes Neurais 基于神经网络分析的Black Litterman投资组合

Learning and Nonlinear Models Pub Date : 2020-12-31 DOI: 10.21528/LNLM-VOL18-NO1-ART5

Diego Guerreiro Bernardes, O. L. Costa

引用次数: 0

Aerial Human Activity Recognition Through a Cognitive Architecture and a New Automata Proposal 基于认知架构和自动机的空中人类活动识别

Learning and Nonlinear Models Pub Date : 2020-09-01 DOI: 10.21528/lnlm-vol18-no1-art1

M. Pinto, Aurelio G. Melo, A. Marcato, C. Moraes

{"title":"Aerial Human Activity Recognition Through a Cognitive Architecture and a New Automata Proposal","authors":"M. Pinto, Aurelio G. Melo, A. Marcato, C. Moraes","doi":"10.21528/lnlm-vol18-no1-art1","DOIUrl":"https://doi.org/10.21528/lnlm-vol18-no1-art1","url":null,"abstract":"Video surveillance often involves several actors in multiple interactions and modelling complex activities becomes a challenge, especially in real environments. When applying autonomous video surveillance, object recognition techniques are used to produce symbolic information related to the information present in a scene. An automaton is a specialized structure capable of accepting or rejecting those symbols producing an efficient computation structure for these types of data processing. This research work presents an innovative structure for the well-known Weighted Automata to organize the information from sensors, grouping these measurements into a symbolic representation of actions that are happening in the real world. This work also proposes a hierarchical architecture formed by a multilevel sensorial system comprised of low, middle and high levels to perceive the environment and to comprehend scenes. The proposed architecture is designed to operate in a decentralized way and onboard of the Unmanned Aerial Vehicles (UAVs). The experiments showed that this system updated effectively the semantic structure given the sequence of information and demonstrated the automaton and architecture effectiveness..","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134310767","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

Distributed Adaptive Filtering on Wireless Sensor Networks with Shared Medium Competition 共享介质竞争无线传感器网络的分布式自适应滤波

Learning and Nonlinear Models Pub Date : 2020-09-01 DOI: 10.21528/lnlm-vol18-no1-art2

Rafael M. Carmo, Luís Tarrataca, J. Colares, F. Henriques, D. B. Haddad, Raphael M. Guedes

{"title":"Distributed Adaptive Filtering on Wireless Sensor Networks with Shared Medium Competition","authors":"Rafael M. Carmo, Luís Tarrataca, J. Colares, F. Henriques, D. B. Haddad, Raphael M. Guedes","doi":"10.21528/lnlm-vol18-no1-art2","DOIUrl":"https://doi.org/10.21528/lnlm-vol18-no1-art2","url":null,"abstract":"Wireless Sensor Networks (WSN) are of significant importance with increasingly diverse and viable applications. They gained even more traction after the IEEE 802.15.4 standard was defined. Distributed adaptive filtering algorithms have added statistical inference to WSN applications, employing techniques that extract data from distributed devices. In contrast, most adaptive filtering contributions do not consider realistic features of the subjacent telecommunications network protocols. Similarly, the telecommunications area typically does not take into account interesting abilities of adaptive filtering algorithms. In this paper, we explore this gap between the two study areas, allowing the development of network-protocol-aware distributed adaptive filtering techniques. In order to explore network realistic behaviors, this paper focuses on distributed inference problems. More specifically, we propose two new diffuse adaptive algorithms, aware of the characteristics of the Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) protocol, namely: (i) Variant Reuse of Coefficients Least Mean Squares (VRCLMS) algorithm; and the (ii) Reuse of Coefficients Least Mean Squares (RC-LMS) algorithm in the Adapt-Then-Combine (ATC) modality. These two new algorithms will bring some advantages, specifically when information is delayed because of too much packet loss. Another advantage will be the addition of the spatial information diversity contribution in the VRC-LMS algorithm.","PeriodicalId":386768,"journal":{"name":"Learning and Nonlinear Models","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130136574","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

Soft Biometrics Classication in Videos Using Transfer Learning and Bidirectional Long Short-Term Memory Networks 基于迁移学习和双向长短期记忆网络的视频软生物识别分类

Learning and Nonlinear Models Pub Date : 2020-09-01 DOI: 10.21528/lnlm-vol18-no1-art4

M. Romero, M. Gutoski, L. T. Hattori, Manassés Ribeiro, H. S. Lopes

引用次数: 1