Hierarchical multi-label classification using local neural networks

Hierarchical multi-label classification is a complex classification task where the classes involved in the problem are hierarchically structured and each example may simultaneously belong to more than one class in each hierarchical level. In this paper, we extend our previous works, where we investigated a new local-based classification method that incrementally trains a multi-layer perceptron for each level of the classification hierarchy. Predictions made by a neural network in a given level are used as inputs to the neural network responsible for the prediction in the next level. We compare the proposed method with one state-of-the-art decision-tree induction method and two decision-tree induction methods, using several hierarchical multi-label classification datasets. We perform a thorough experimental analysis, showing that our method obtains competitive results to a robust global method regarding both precision and recall evaluation measures.

Nondestructive evaluation of concrete compression strength by means of Artificial Neural Network (ANN)

The evaluation of structural performance of existing concrete buildings, built according to standards and materials quite different to those available today, requires procedures and methods able to cover lack of data about mechanical material properties and reinforcement detailing. To this end detailed inspections and test on materials are required. As a consequence tests on drilled cores are required; on the other end, it is stated that non-destructive testing (NDT) cannot be used as the only mean to get structural information, but can be used in conjunction with destructive testing (DT) by a representative correlation between DT and NDT. The aim of this study is to verify the accuracy of some formulas of correlation available in literature between measured parameters, i.e. rebound index, ultrasonic pulse velocity and compressive strength (SonReb Method). To this end a relevant number of DT and NDT tests has been performed on many school buildings located in Cesena (Italy). The above relationships have been assessed on site correlating NDT results to strength of core drilled in adjacent locations. Nevertheless, concrete compressive strength assessed by means of NDT methods and evaluated with correlation formulas has the advantage of being able to be implemented and used for future applications in a much more simple way than other methods, even if its accuracy is strictly limited to the analysis of concretes having the same characteristics as those used for their calibration. This limitation warranted a search for a different evaluation method for the non-destructive parameters obtained on site. To this aim, the methodology of neural identification of compressive strength is presented. Artificial Neural Network (ANN) suitable for the specific analysis were chosen taking into account the development presented in the literature in this field. The networks were trained and tested in order to detect a more reliable strength identification methodology.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Neural Networks as Cybernetic Systems - Part I

Neural Networks as Cybernetic Systems is a textbox that combines classical systems theory with artificial neural network technology.

Neural Networks as Cybernetic Systems - Part II

Neural Networks as Cybernetic Systems is a textbox that combines classical systems theory with artificial neural network technology.

Neural Networks as Cybernetic Systems (3rd and revised edition) - Part I

Neural Networks as Cybernetic Systems is a textbox that combines classical systems theory with artificial neural network technology. This third edition essentially compares with the 2nd one, but has been improved by correction of errors and by a rearrangement and minor expansion of the sections referring to recurrent networks. These changes hopefully allow for an easier comprehension of the essential aspects of this important domain that has received growing attention during the last years.

Neural Networks as Cybernetic Systems (3rd and revised edition) - Part 2

eural Networks as Cybernetic Systems is a textbox that combines classical systems theory with artificial neural network technology. This third edition essentially compares with the 2nd one, but has been improved by correction of errors and by a rearrangement and minor expansion of the sections referring to recurrent networks. These changes hopefully allow for an easier comprehension of the essential aspects of this important domain that has received growing attention during the last years.

Aplicación de técnicas de inteligencia artificial a la predicción de contaminantes atmosféricos

Abstract Air pollution is a big threat and a phenomenon that has a specific impact on human health, in addition, changes that occur in the chemical composition of the atmosphere can change the weather and cause acid rain or ozone destruction. Those are phenomena of global importance. The World Health Organization (WHO) considerates air pollution as one of the most important global priorities. Salamanca, Gto., Mexico has been ranked as one of the most polluted cities in this country. The industry of the area led to a major economic development and rapid population growth in the second half of the twentieth century. The impact in the air quality is important and significant efforts have been made to measure the concentrations of pollutants. The main pollution sources are locally based plants in the chemical and power generation sectors. The registered concerning pollutants are Sulphur Dioxide (SO2) and particles on the order of ∼10 micrometers or less (PM10). The prediction in the concentration of those pollutants can be a powerful tool in order to take preventive measures such as the reduction of emissions and alerting the affected population. In this PhD thesis we propose a model to predict concentrations of pollutants SO2 and PM10 for each monitoring booth in the Atmospheric Monitoring Network Salamanca (REDMAS - for its spanish acronym). The proposed models consider the use of meteorological variables as factors influencing the concentration of pollutants. The information used along this work is the current real data from REDMAS. In the proposed model, Artificial Neural Networks (ANN) combined with clustering algorithms are used. The type of ANN used is the Multilayer Perceptron with a hidden layer, using separate structures for the prediction of each pollutant. The meteorological variables used for prediction were: Wind Direction (WD), wind speed (WS), Temperature (T) and relative humidity (RH). Clustering algorithms, K-means and Fuzzy C-means, are used to find relationships between air pollutants and weather variables under consideration, which are added as input of the RNA. Those relationships provide information to the ANN in order to obtain the prediction of the pollutants. The results of the model proposed in this work are compared with the results of a multivariate linear regression and multilayer perceptron neural network. The evaluation of the prediction is calculated with the mean absolute error, the root mean square error, the correlation coefficient and the index of agreement. The results show the importance of meteorological variables in the prediction of the concentration of the pollutants SO2 and PM10 in the city of Salamanca, Gto., Mexico. The results show that the proposed model perform better than multivariate linear regression and multilayer perceptron neural network. The models implemented for each monitoring booth have the ability to make predictions of air quality that can be used in a system of real-time forecasting and human health impact analysis. Among the main results of the development of this thesis we can cite: A model based on artificial neural network combined with clustering algorithms for prediction with a hour ahead of the concentration of each pollutant (SO2 and PM10) is proposed. A different model was designed for each pollutant and for each of the three monitoring booths of the REDMAS. A model to predict the average of pollutant concentration in the next 24 hours of pollutants SO2 and PM10 is proposed, based on artificial neural network combined with clustering algorithms. Model was designed for each booth of the REDMAS and each pollutant separately. Resumen La contaminación atmosférica es una amenaza aguda, constituye un fenómeno que tiene particular incidencia sobre la salud del hombre. Los cambios que se producen en la composición química de la atmósfera pueden cambiar el clima, producir lluvia ácida o destruir el ozono, fenómenos todos ellos de una gran importancia global. La Organización Mundial de la Salud (OMS) considera la contaminación atmosférica como una de las más importantes prioridades mundiales. Salamanca, Gto., México; ha sido catalogada como una de las ciudades más contaminadas en este país. La industria de la zona propició un importante desarrollo económico y un crecimiento acelerado de la población en la segunda mitad del siglo XX. Las afectaciones en el aire son graves y se han hecho importantes esfuerzos por medir las concentraciones de los contaminantes. Las principales fuentes de contaminación son fuentes fijas como industrias químicas y de generación eléctrica. Los contaminantes que se han registrado como preocupantes son el Bióxido de Azufre (SO2) y las Partículas Menores a 10 micrómetros (PM10). La predicción de las concentraciones de estos contaminantes puede ser una potente herramienta que permita tomar medidas preventivas como reducción de emisiones a la atmósfera y alertar a la población afectada. En la presente tesis doctoral se propone un modelo de predicción de concentraci ón de los contaminantes más críticos SO2 y PM10 para cada caseta de monitorización de la Red de Monitorización Atmosférica de Salamanca (REDMAS). Los modelos propuestos plantean el uso de las variables meteorol ógicas como factores que influyen en la concentración de los contaminantes. La información utilizada durante el desarrollo de este trabajo corresponde a datos reales obtenidos de la REDMAS. En el Modelo Propuesto (MP) se aplican Redes Neuronales Artificiales (RNA) combinadas con algoritmos de agrupamiento. La RNA utilizada es el Perceptrón Multicapa con una capa oculta, utilizando estructuras independientes para la predicción de cada contaminante. Las variables meteorológicas disponibles para realizar la predicción fueron: Dirección de Viento (DV), Velocidad de Viento (VV), Temperatura (T) y Humedad Relativa (HR). Los algoritmos de agrupamiento K-means y Fuzzy C-means son utilizados para encontrar relaciones existentes entre los contaminantes atmosféricos en estudio y las variables meteorológicas. Dichas relaciones aportan información a las RNA para obtener la predicción de los contaminantes, la cual es agregada como entrada de las RNA. Los resultados del modelo propuesto en este trabajo son comparados con los resultados de una Regresión Lineal Multivariable (RLM) y un Perceptrón Multicapa (MLP). La evaluación de la predicción se realiza con el Error Medio Absoluto, la Raíz del Error Cuadrático Medio, el coeficiente de correlación y el índice de acuerdo. Los resultados obtenidos muestran la importancia de las variables meteorológicas en la predicción de la concentración de los contaminantes SO2 y PM10 en la ciudad de Salamanca, Gto., México. Los resultados muestran que el MP predice mejor la concentración de los contaminantes SO2 y PM10 que los modelos RLM y MLP. Los modelos implementados para cada caseta de monitorizaci ón tienen la capacidad para realizar predicciones de calidad del aire, estos modelos pueden ser implementados en un sistema que permita realizar la predicción en tiempo real y analizar el impacto en la salud de la población. Entre los principales resultados obtenidos del desarrollo de esta tesis podemos citar: Se propone un modelo basado en una red neuronal artificial combinado con algoritmos de agrupamiento para la predicción con una hora de anticipaci ón de la concentración de cada contaminante (SO2 y PM10). Se diseñó un modelo diferente para cada contaminante y para cada una de las tres casetas de monitorización de la REDMAS. Se propone un modelo de predicción del promedio de la concentración de las próximas 24 horas de los contaminantes SO2 y PM10, basado en una red neuronal artificial combinado con algoritmos de agrupamiento. Se diseñó un modelo para cada caseta de monitorización de la REDMAS y para cada contaminante por separado.

Artificial Neural Nets with Interaction of Afferents

The aim is to obtain computationally more powerful, neuro physiologically founded, artificial neurons and neural nets. Artificial Neural Nets (ANN) of the Perceptron type evolved from the original proposal by McCulloch an Pitts classical paper [1]. Essentially, they keep the computing structure of a linear machine followed by a non linear operation. The McCulloch-Pitts formal neuron (which was never considered by the author’s to be models of real neurons) consists of the simplest case of a linear computation of the inputs followed by a threshold. Networks of one layer cannot compute anylogical function of the inputs, but only those which are linearly separable. Thus, the simple exclusive OR (contrast detector) function of two inputs requires two layers of formal neurons

On the Biological Plausibility of Artificial Metaplasticity

The training algorithm studied in this paper is inspired by the biological metaplasticity property of neurons. Tested on different multidisciplinary applications, it achieves a more efficient training and improves Artificial Neural Network Performance. The algorithm has been recently proposed for Artificial Neural Networks in general, although for the purpose of discussing its biological plausibility, a Multilayer Perceptron has been used. During the training phase, the artificial metaplasticity multilayer perceptron could be considered a new probabilistic version of the presynaptic rule, as during the training phase the algorithm assigns higher values for updating the weights in the less probable activations than in the ones with higher probability

Polynomial approximation using particle swarm optimization of lineal enhanced neural networks with no hidden layers.

This paper presents some ideas about a new neural network architecture that can be compared to a Taylor analysis when dealing with patterns. Such architecture is based on lineal activation functions with an axo-axonic architecture. A biological axo-axonic connection between two neurons is defined as the weight in a connection in given by the output of another third neuron. This idea can be implemented in the so called Enhanced Neural Networks in which two Multilayer Perceptrons are used; the first one will output the weights that the second MLP uses to computed the desired output. This kind of neural network has universal approximation properties even with lineal activation functions. There exists a clear difference between cooperative and competitive strategies. The former ones are based on the swarm colonies, in which all individuals share its knowledge about the goal in order to pass such information to other individuals to get optimum solution. The latter ones are based on genetic models, that is, individuals can die and new individuals are created combining information of alive one; or are based on molecular/celular behaviour passing information from one structure to another. A swarm-based model is applied to obtain the Neural Network, training the net with a Particle Swarm algorithm.

Artificial analysis of molecular marker loci linked to tree resistance response by an artificial neural network

One of the biggest challenges that software developers face is to make an accurate estimate of the project effort. Radial basis function neural networks have been used to software effort estimation in this work using NASA dataset. This paper evaluates and compares radial basis function versus a regression model. The results show that radial basis function neural network have obtained less Mean Square Error than the regression method.

Aplicação de uma rede neural artificial para a avaliação da rugosidade e soprosidade vocal

A avaliação perceptivo-auditiva tem papel fundamental no estudo e na avaliação da voz, no entanto, por ser subjetiva está sujeita a imprecisões e variações. Por outro lado, a análise acústica permite a reprodutibilidade de resultados, porém precisa ser aprimorada, pois não analisa com precisão vozes com disfonias mais intensas e com ondas caóticas. Assim, elaborar medidas que proporcionem conhecimentos confiáveis em relação à função vocal resulta de uma necessidade antiga dentro desta linha de pesquisa e atuação clínica. Neste contexto, o uso da inteligência artificial, como as redes neurais artificiais, indica ser uma abordagem promissora. Objetivo: Validar um sistema automático utilizando redes neurais artificiais para a avaliação de vozes rugosas e soprosas. Materiais e métodos: Foram selecionadas 150 vozes, desde neutras até com presença em grau intenso de rugosidade e/ou soprosidade, do banco de dados da Clínica de Fonoaudiologia da Faculdade de Odontologia de Bauru (FOB/USP). Dessas vozes, 23 foram excluídas por não responderem aos critérios de inclusão na amostra, assim utilizaram-se 123 vozes. Procedimentos: avaliação perceptivo-auditiva pela escala visual analógica de 100 mm e pela escala numérica de quatro pontos; extração de características do sinal de voz por meio da Transformada Wavelet Packet e dos parâmetros acústicos: jitter, shimmer, amplitude da derivada e amplitude do pitch; e validação do classificador por meio da parametrização, treino, teste e avaliação das redes neurais artificiais. Resultados: Na avaliação perceptivo-auditiva encontrou-se, por meio do teste Coeficiente de Correlação Intraclasse (CCI), concordâncias inter e intrajuiz excelentes, com p = 0,85 na concordância interjuízes e p variando de 0,87 a 0,93 nas concordâncias intrajuiz. Em relação ao desempenho da rede neural artificial, na discriminação da soprosidade e da rugosidade e dos seus respectivos graus, encontrou-se o melhor desempenho para a soprosidade no subconjunto composto pelo jitter, amplitude do pitch e frequência fundamental, no qual obteve-se taxa de acerto de 74%, concordância excelente com a avaliação perceptivo-auditiva da escala visual analógica (0,80 no CCI) e erro médio de 9 mm. Para a rugosidade, o melhor subconjunto foi composto pela Transformada Wavelet Packet com 1 nível de decomposição, jitter, shimmer, amplitude do pitch e frequência fundamental, no qual obteve-se 73% de acerto, concordância excelente (0,84 no CCI), e erro médio de 10 mm. Conclusão: O uso da inteligência artificial baseado em redes neurais artificiais na identificação, e graduação da rugosidade e da soprosidade, apresentou confiabilidade excelente (CCI > 0,80), com resultados semelhantes a concordância interjuízes. Dessa forma, a rede neural artificial revela-se como uma metodologia promissora de avaliação vocal, tendo sua maior vantagem a objetividade na avaliação.

Designing Regularizers and Architectures for Recurrent Neural Networks

Cette thèse contribue a la recherche vers l'intelligence artificielle en utilisant des méthodes connexionnistes. Les réseaux de neurones récurrents sont un ensemble de modèles séquentiels de plus en plus populaires capable en principe d'apprendre des algorithmes arbitraires. Ces modèles effectuent un apprentissage en profondeur, un type d'apprentissage machine. Sa généralité et son succès empirique en font un sujet intéressant pour la recherche et un outil prometteur pour la création de l'intelligence artificielle plus générale. Le premier chapitre de cette thèse donne un bref aperçu des sujets de fonds: l'intelligence artificielle, l'apprentissage machine, l'apprentissage en profondeur et les réseaux de neurones récurrents. Les trois chapitres suivants couvrent ces sujets de manière de plus en plus spécifiques. Enfin, nous présentons quelques contributions apportées aux réseaux de neurones récurrents. Le chapitre \ref{arxiv1} présente nos travaux de régularisation des réseaux de neurones récurrents. La régularisation vise à améliorer la capacité de généralisation du modèle, et joue un role clé dans la performance de plusieurs applications des réseaux de neurones récurrents, en particulier en reconnaissance vocale. Notre approche donne l'état de l'art sur TIMIT, un benchmark standard pour cette tâche. Le chapitre \ref{cpgp} présente une seconde ligne de travail, toujours en cours, qui explore une nouvelle architecture pour les réseaux de neurones récurrents. Les réseaux de neurones récurrents maintiennent un état caché qui représente leurs observations antérieures. L'idée de ce travail est de coder certaines dynamiques abstraites dans l'état caché, donnant au réseau une manière naturelle d'encoder des tendances cohérentes de l'état de son environnement. Notre travail est fondé sur un modèle existant; nous décrivons ce travail et nos contributions avec notamment une expérience préliminaire.

Designing Regularizers and Architectures for Recurrent Neural Networks

Cette thèse contribue a la recherche vers l'intelligence artificielle en utilisant des méthodes connexionnistes. Les réseaux de neurones récurrents sont un ensemble de modèles séquentiels de plus en plus populaires capable en principe d'apprendre des algorithmes arbitraires. Ces modèles effectuent un apprentissage en profondeur, un type d'apprentissage machine. Sa généralité et son succès empirique en font un sujet intéressant pour la recherche et un outil prometteur pour la création de l'intelligence artificielle plus générale. Le premier chapitre de cette thèse donne un bref aperçu des sujets de fonds: l'intelligence artificielle, l'apprentissage machine, l'apprentissage en profondeur et les réseaux de neurones récurrents. Les trois chapitres suivants couvrent ces sujets de manière de plus en plus spécifiques. Enfin, nous présentons quelques contributions apportées aux réseaux de neurones récurrents. Le chapitre \ref{arxiv1} présente nos travaux de régularisation des réseaux de neurones récurrents. La régularisation vise à améliorer la capacité de généralisation du modèle, et joue un role clé dans la performance de plusieurs applications des réseaux de neurones récurrents, en particulier en reconnaissance vocale. Notre approche donne l'état de l'art sur TIMIT, un benchmark standard pour cette tâche. Le chapitre \ref{cpgp} présente une seconde ligne de travail, toujours en cours, qui explore une nouvelle architecture pour les réseaux de neurones récurrents. Les réseaux de neurones récurrents maintiennent un état caché qui représente leurs observations antérieures. L'idée de ce travail est de coder certaines dynamiques abstraites dans l'état caché, donnant au réseau une manière naturelle d'encoder des tendances cohérentes de l'état de son environnement. Notre travail est fondé sur un modèle existant; nous décrivons ce travail et nos contributions avec notamment une expérience préliminaire.