On Recurrent and Deep Neural Networks

L'apprentissage profond est un domaine de recherche en forte croissance en apprentissage automatique qui est parvenu à des résultats impressionnants dans différentes tâches allant de la classification d'images à la parole, en passant par la modélisation du langage. Les réseaux de neurones récurrents, une sous-classe d'architecture profonde, s'avèrent particulièrement prometteurs. Les réseaux récurrents peuvent capter la structure temporelle dans les données. Ils ont potentiellement la capacité d'apprendre des corrélations entre des événements éloignés dans le temps et d'emmagasiner indéfiniment des informations dans leur mémoire interne. Dans ce travail, nous tentons d'abord de comprendre pourquoi la profondeur est utile. Similairement à d'autres travaux de la littérature, nos résultats démontrent que les modèles profonds peuvent être plus efficaces pour représenter certaines familles de fonctions comparativement aux modèles peu profonds. Contrairement à ces travaux, nous effectuons notre analyse théorique sur des réseaux profonds acycliques munis de fonctions d'activation linéaires par parties, puisque ce type de modèle est actuellement l'état de l'art dans différentes tâches de classification. La deuxième partie de cette thèse porte sur le processus d'apprentissage. Nous analysons quelques techniques d'optimisation proposées récemment, telles l'optimisation Hessian free, la descente de gradient naturel et la descente des sous-espaces de Krylov. Nous proposons le cadre théorique des méthodes à région de confiance généralisées et nous montrons que plusieurs de ces algorithmes développés récemment peuvent être vus dans cette perspective. Nous argumentons que certains membres de cette famille d'approches peuvent être mieux adaptés que d'autres à l'optimisation non convexe. La dernière partie de ce document se concentre sur les réseaux de neurones récurrents. Nous étudions d'abord le concept de mémoire et tentons de répondre aux questions suivantes: Les réseaux récurrents peuvent-ils démontrer une mémoire sans limite? Ce comportement peut-il être appris? Nous montrons que cela est possible si des indices sont fournis durant l'apprentissage. Ensuite, nous explorons deux problèmes spécifiques à l'entraînement des réseaux récurrents, à savoir la dissipation et l'explosion du gradient. Notre analyse se termine par une solution au problème d'explosion du gradient qui implique de borner la norme du gradient. Nous proposons également un terme de régularisation conçu spécifiquement pour réduire le problème de dissipation du gradient. Sur un ensemble de données synthétique, nous montrons empiriquement que ces mécanismes peuvent permettre aux réseaux récurrents d'apprendre de façon autonome à mémoriser des informations pour une période de temps indéfinie. Finalement, nous explorons la notion de profondeur dans les réseaux de neurones récurrents. Comparativement aux réseaux acycliques, la définition de profondeur dans les réseaux récurrents est souvent ambiguë. Nous proposons différentes façons d'ajouter de la profondeur dans les réseaux récurrents et nous évaluons empiriquement ces propositions.

“Reconstruction of Gene Regulatory Network from Expression Profile of Plasma RNA Data of Colorectal Cancer Patients using Soft Computing Techniques”

Microarray data analysis is one of data mining tool which is used to extract meaningful information hidden in biological data. One of the major focuses on microarray data analysis is the reconstruction of gene regulatory network that may be used to provide a broader understanding on the functioning of complex cellular systems. Since cancer is a genetic disease arising from the abnormal gene function, the identification of cancerous genes and the regulatory pathways they control will provide a better platform for understanding the tumor formation and development. The major focus of this thesis is to understand the regulation of genes responsible for the development of cancer, particularly colorectal cancer by analyzing the microarray expression data. In this thesis, four computational algorithms namely fuzzy logic algorithm, modified genetic algorithm, dynamic neural fuzzy network and Takagi Sugeno Kang-type recurrent neural fuzzy network are used to extract cancer specific gene regulatory network from plasma RNA dataset of colorectal cancer patients. Plasma RNA is highly attractive for cancer analysis since it requires a collection of small amount of blood and it can be obtained at any time in repetitive fashion allowing the analysis of disease progression and treatment response.

Using Recurrent Networks for Dimensionality Reduction

This report explores how recurrent neural networks can be exploited for learning high-dimensional mappings. Since recurrent networks are as powerful as Turing machines, an interesting question is how recurrent networks can be used to simplify the problem of learning from examples. The main problem with learning high-dimensional functions is the curse of dimensionality which roughly states that the number of examples needed to learn a function increases exponentially with input dimension. This thesis proposes a way of avoiding this problem by using a recurrent network to decompose a high-dimensional function into many lower dimensional functions connected in a feedback loop.

Inverse model control using recurrent networks

This paper illustrates how internal model control of nonlinear processes can be achieved by recurrent neural networks, e.g. fully connected Hopfield networks. It is shown that using results developed by Kambhampati et al. (1995), that once a recurrent network model of a nonlinear system has been produced, a controller can be produced which consists of the network comprising the inverse of the model and a filter. Thus, the network providing control for the nonlinear system does not require any training after it has been trained to model the nonlinear system. Stability and other issues of importance for nonlinear control systems are also discussed.

The relative order and inverses of recurrent networks

Differential geometry is used to investigate the structure of neural-network-based control systems. The key aspect is relative order—an invariant property of dynamic systems. Finite relative order allows the specification of a minimal architecture for a recurrent network. Any system with finite relative order has a left inverse. It is shown that a recurrent network with finite relative order has a local inverse that is also a recurrent network with the same weights. The results have implications for the use of recurrent networks in the inverse-model-based control of nonlinear systems.

Relative order defines a topology for recurrent networks

This paper uses techniques from control theory in the analysis of trained recurrent neural networks. Differential geometry is used as a framework, which allows the concept of relative order to be applied to neural networks. Any system possessing finite relative order has a left-inverse. Any recurrent network with finite relative order also has an inverse, which is shown to be a recurrent network.

A review of online learning in supervised neural networks

Learning in neural networks can broadly be divided into two categories, viz., off-line (or batch) learning and online (or incremental) learning. In this paper, a review of a variety of supervised neural networks with online learning capabilities is presented. Specifically, we focus on articles published in main indexed journals in the past 10 years (2003–2013). We examine a number of key neural network architectures, which include feedforward neural networks, recurrent neural networks, fuzzy neural networks, and other related networks. How the online learning methodologies are incorporated into these networks is exemplified, and how they are applied to solving problems in different domains is highlighted. A summary of the review that covers different network architectures and their applications is presented.

Neural approach for solving several types of optimization problems

Neural networks consist of highly interconnected and parallel nonlinear processing elements that are shown to be extremely effective in computation. This paper presents an architecture of recurrent neural net-works that can be used to solve several classes of optimization problems. More specifically, a modified Hopfield network is developed and its inter-nal parameters are computed explicitly using the valid-subspace technique. These parameters guarantee the convergence of the network to the equilibrium points, which represent a solution of the problem considered. The problems that can be treated by the proposed approach include combinatorial optimiza-tion problems, dynamic programming problems, and nonlinear optimization problems.

Projeto E análise de uma rede neural para resolver problemas de programação dinâmica

Systems based on artificial neural networks have high computational rates due to the use of a massive number of simple processing elements and the high degree of connectivity between these elements. Neural networks with feedback connections provide a computing model capable of solving a large class of optimization problems. This paper presents a novel approach for solving dynamic programming problems using artificial neural networks. More specifically, a modified Hopfield network is developed and its internal parameters are computed using the valid-subspace technique. These parameters guarantee the convergence of the network to the equilibrium points which represent solutions (not necessarily optimal) for the dynamic programming problem. Simulated examples are presented and compared with other neural networks. The results demonstrate that proposed method gives a significant improvement.

Estimation of electrical machine speed using sensorless technology and neural networks

The use of sensorless technologies is an increasing tendency on industrial drivers for electrical machines. The estimation of electrical and mechanical parameters involved with the electrical machine control is used very frequently in order to avoid measurement of all variables related to this process. The cost reduction may also be considered in industrial drivers, besides the increasing robustness of the system, as an advantage of the use of sensorless technologies. This work proposes the use of a recurrent artificial neural network to estimate the speed of induction motor for sensorless control schemes using one single current sensor. Simulation and experimental results are presented to validate the proposed approach. ©2008 IEEE.

Detecção de refletores sísmicos por rede neural discreta

As redes neurais artificiais têm provado serem uma poderosa técnica na resolução de uma grande variedade de problemas de otimização. Nesta dissertação é desenvolvida uma nova rede neural, tipo recorrente, sem realimentação (self-feedback loops) e sem neurônios ocultos, para o processamento do sinal sísmico, para fornecer a posição temporal, a polaridade e as amplitudes estimadas dos refletores sísmicos, representadas pelos seus coeficientes de reflexão. A principal característica dessa nova rede neural consiste no tipo de função de ativação utilizada, a qual permite três possíveis estados para o neurônio. Busca-se estimar a posição dos refletores sísmicos e reproduzir as verdadeiras polaridades desses refletores. A idéia básica desse novo tipo de rede, aqui denominada rede neural discreta (RND), é relacionar uma função objeto, que descreve o problema geofísico, com a função de Liapunov, que descreve a dinâmica da rede neural. Deste modo, a dinâmica da rede leva a uma minimização local da sua função de Liapunov e consequentemente leva a uma minimização da função objeto. Assim, com uma codificação conveniente do sinal de saída da rede tem-se uma solução do problema geofísico. A avaliação operacional da arquitetura desta rede neural artificial é realizada em dados sintéticos gerados através do modelo convolucional simples e da teoria do raio. A razão é para explicar o comportamento da rede com dados contaminados por ruído, e diante de pulsos fonte de fases mínima, máxima e misturada.

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.

Glucose-Insulin regulator for type 1 diabetes using high order neural networks

In this paper a Glucose-Insulin regulator for Type 1 Diabetes using artificial neural networks (ANN) is proposed. This is done using a discrete recurrent high order neural network in order to identify and control a nonlinear dynamical system which represents the pancreas? beta-cells behavior of a virtual patient. The ANN which reproduces and identifies the dynamical behavior system, is configured as series parallel and trained on line using the extended Kalman filter algorithm to achieve a quickly convergence identification in silico. The control objective is to regulate the glucose-insulin level under different glucose inputs and is based on a nonlinear neural block control law. A safety block is included between the control output signal and the virtual patient with type 1 diabetes mellitus. Simulations include a period of three days. Simulation results are compared during the overnight fasting period in Open-Loop (OL) versus Closed- Loop (CL). Tests in Semi-Closed-Loop (SCL) are made feedforward in order to give information to the control algorithm. We conclude the controller is able to drive the glucose to target in overnight periods and the feedforward is necessary to control the postprandial period.

Generalizing of Neural Nets: Functional Nets of Special Type

Special generalizing for the artificial neural nets: so called RFT – FN – is under discussion in the report. Such refinement touch upon the constituent elements for the conception of artificial neural network, namely, the choice of main primary functional elements in the net, the way to connect them(topology) and the structure of the net as a whole. As to the last, the structure of the functional net proposed is determined dynamically just in the constructing the net by itself by the special recurrent procedure. The number of newly joining primary functional elements, the topology of its connecting and tuning of the primary elements is the content of the each recurrent step. The procedure is terminated under fulfilling “natural” criteria relating residuals for example. The functional proposed can be used in solving the approximation problem for the functions, represented by its observations, for classifying and clustering, pattern recognition, etc. Recurrent procedure provide for the versatile optimizing possibilities: as on the each step of the procedure and wholly: by the choice of the newly joining elements, topology, by the affine transformations if input and intermediate coordinate as well as by its nonlinear coordinate wise transformations. All considerations are essentially based, constructively and evidently represented by the means of the Generalized Inverse.