Activation function study for wavelet network

The main purpose of this paper is to investigate theoretically and experimentally the use of family of Polynomial Powers of the Sigmoid (PPS) Function Networks applied in speech signal representation and function approximation. This paper carries out practical investigations in terms of approximation fitness (LSE), time consuming (CPU Time), computational complexity (FLOP) and representation power (Number of Activation Function) for different PPS activation functions. We expected that different activation functions can provide performance variations and further investigations will guide us towards a class of mappings associating the best activation function to solve a class of problems under certain criteria.

Approximation of hyperbolic tangent activation function using hybrid methods

Artificial Neural Networks are widely used in various applications in engineering, as such solutions of nonlinear problems. The implementation of this technique in reconfigurable devices is a great challenge to researchers by several factors, such as floating point precision, nonlinear activation function, performance and area used in FPGA. The contribution of this work is the approximation of a nonlinear function used in ANN, the popular hyperbolic tangent activation function. The system architecture is composed of several scenarios that provide a tradeoff of performance, precision and area used in FPGA. The results are compared in different scenarios and with current literature on error analysis, area and system performance. © 2013 IEEE.

Multinodal Load Forecasting in Power Electric Systems using a Neural Network with Radial Basis Function

In this paper we present the results of the use of a methodology for multinodal load forecasting through an artificial neural network-type Multilayer Perceptron, making use of radial basis functions as activation function and the Backpropagation algorithm, as an algorithm to train the network. This methodology allows you to make the prediction at various points in power system, considering different types of consumers (residential, commercial, industrial) of the electric grid, is applied to the problem short-term electric load forecasting (24 hours ahead). We use a database (Centralised Dataset - CDS) provided by the Electricity Commission de New Zealand to this work.

Feedforward neural networks based on PPS-wavelet activation functions

Function approximation is a very important task in environments where computation has to be based on extracting information from data samples in real world processes. Neural networks and wavenets have been recently seen as attractive tools for developing efficient solutions for many real world problems in function approximation. In this paper, it is shown how feedforward neural networks can be built using a different type of activation function referred to as the PPS-wavelet. An algorithm is presented to generate a family of PPS-wavelets that can be used to efficiently construct feedforward networks for function approximation.

Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks

Wavelet functions have been used as the activation function in feedforward neural networks. An abundance of R&D has been produced on wavelet neural network area. Some successful algorithms and applications in wavelet neural network have been developed and reported in the literature. However, most of the aforementioned reports impose many restrictions in the classical backpropagation algorithm, such as low dimensionality, tensor product of wavelets, parameters initialization, and, in general, the output is one dimensional, etc. In order to remove some of these restrictions, a family of polynomial wavelets generated from powers of sigmoid functions is presented. We described how a multidimensional wavelet neural networks based on these functions can be constructed, trained and applied in pattern recognition tasks. As an example of application for the method proposed, it is studied the exclusive-or (XOR) problem.

Computer-aided design of a novel ligand for retinoic acid receptor in cancer chemotherapy

The isotypes of RAR and RXR are retinoic acid and retinoid X acid receptors, respectively, whose ligand-binding domain contains the ligand-dependent activation function, with distinct pharmacological targets for retinoids, involved in the treatment of various cancers and skin diseases. Due to the major challenge which cancer treatment and cure still imposes after many decades to the international scientific community, there is actually considerable interest in new ligands with increased bioactivity. We have focused on the retinoid acid receptor, which is considered an interesting target for drug design. In this work, we carried out density functional geometry optimizations, and different docking procedures. We performed screening in a large database (hundreds of thousands of molecules which we optimized at the AM1 level) yielding a set of potential bioactive ligands. A new ligand was selected and optimized at the B3LYP/6-31G* level. A flexible docking program was used to investigate the interactions between the receptor and the new ligand. The result of this work is compared with several crystallographic ligands of RAR. Our theoretically more bioactive new-ligand indicates stronger and more hydrogen bonds as well as hydrophobic interactions with the receptor. (c) 2005 Wiley Periodicals, Inc.

Identificação e controle de sistemas dinâmicos utilizando redes wavelets

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Desenvolvimento de redes neurais para previsão de cargas elétricas de sistemas de energia elétrica

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Projeto e implementação de redes neurais artificiais em distintos níveis de abstrações para o reconhecimento de deficiências de diversos macronutrientes e cultivares

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Desenvolvimento de uma rede neuro-fuzzy para a previsão da carga

Pós-graduação em Engenharia Elétrica - FEIS

Activation of the brain melanocortin system is required for leptin-induced modulation of chemorespiratory function

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Increased pancreatic islet mass is accompanied by activation of the insulin receptor substrate-2/serine-threonine kinase pathway and augmented cyclin D(2) protein levels in insulin-resistant rats

It is well known that glucocorticoids induce peripheral insulin resistance in rodents and humans. Here, we investigated the structural and ultrastructural modifications, as well as the proteins involved in beta-cell function and proliferation, in islets from insulin-resistant rats. Adult male Wistar rats were made insulin resistant by daily administration of dexamethasone (DEX; 1mg/kg, i.p.) for five consecutive days, whilst control (CTL) rats received saline alone. Structure analyses showed a marked hypertrophy of DEX islets with an increase of 1.7-fold in islet mass and of 1.6-fold in islet density compared with CTL islets (P < 0.05). Ultrastructural evaluation of islets revealed an increased amount of secreting organelles, such as endoplasmic reticulum and Golgi apparatus in DEX islets. Mitotic figures were observed in DEX islets at structural and ultrastructural levels. Beta-cell proliferation, evaluated at the immunohistochemical level using anti-PCNA (proliferating cell nuclear antigen), showed an increase in pancreatic beta-cell proliferation of 6.4-fold in DEX islets compared with CTL islets (P < 0.0001). Increases in insulin receptor substrate-2 (IRS-2), phosphorylated-serine-threonine kinase AKT (p-AKT), cyclin D(2) and a decrease in retinoblastoma protein (pRb) levels were observed in DEX islets compared with CTL islets (P < 0.05). Therefore, during the development of insulin resistance, the endocrine pancreas adapts itself increasing beta-cell mass and proliferation, resulting in an amelioration of the functions. The potential mechanisms that underlie these events involve the activation of the IRS-2/AKT pathway and activation of the cell cycle, mediated by cyclin D(2). These adaptations permit the maintenance of glycaemia at near-physiological ranges.

The Poisson-exponential regression model under different latent activation schemes

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Tobacco smoke-induced left ventricular remodelling is not associated with metalloproteinase-2 or -9 activation

Aim: To investigate the role of MMP-2 and MMP-9 in cardiac remodelling induced by tobacco smoke exposure in rats.Methods: Rats were allocated into two groups: C (n = 9): control animals; ETS (n = 9): exposed to tobacco smoke. After 4months, the animals underwent echocardiography, morphometric study and determination of MMP-2 and MMP-9 activity.Results: ETS rats had larger diastolic (C= 15.6 +/- 1.2 mm/kg, ETS = 18.0 +/- 0.9 mm/kg; p < 0.001) and systolic (C= 7.3 +/- 1.2 mm/kg, ETS = 9.2 0.9 mm/kg; p = 0.001) ventricular diameters adjusted for body weight. Fractional shortening (C= 53 +/- 4.8%, ETS = 48 +/- 3.3%; p = 0.031) and ejection fraction (C= 0. 89 +/- 0.03 5 ETS = 0. 86 +/- 0.02; p = 0.03 0) were smaller in the ETS group. Myocyte cross-sectional area (C= 245 8 mu m(2), ETS=253 8 mu m(2); p = 0.028) was higher in ETS rats. There were no differences in MNtP-2 (C=50 +/- 14%; ETS 43 +/- 11%, p 0.22 +/- 8) or MMP-9 (C=0.36 +/- 0.3%; ETS=0.62 +/- 0.3%, p=0.630) activity between the groups.Conclusion: MMP-2 and MMP-9 did not participate in the remodelling process induced by tobacco smoke exposure. (c) 2007 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.

Platelet activation: ultrastructure and morphometry in platelet-rich plasma of horses

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)