Two-layer Perceptron Research Articles

Multifunctional substrate of Al alloy based on general hierarchical micro/nanostructures: superamphiphobicity and enhanced corrosion resistance.

Aluminum alloys are vulnerable to penetrating and peeling failures in seawater and preparing a barrier coating to isolate the substrate from corrosive medium is an effective anticorrosion method. Inspired by the lotus leaves effect, a wetting alloy surface with enhanced anticorrosion behavior has been prepared via etch, deposition, and low-surface-energy modification. Results indicate that excellent superamphiphobicity has been achieved after the modification of the constructed hierarchical labyrinth-like microstructures and dendritic nanostructures. The as-prepared surface is also found with good chemical stability and mechanical durability. Furthermore, superior anticorrosion behaviors of the resultant samples in seawater are investigated by electrochemical measurements. Due to trapped air in micro/nanostructures, the newly presented solid-air-liquid contacting interface can help to resist the seawater penetration by greatly reducing the interface interaction between corrosive ions and the superamphiphobic surface. Finally, an optimized two-layer perceptron artificial neural network is set up to model and predict the cause-and-effect relationship between preparation conditions and the anticorrosion parameters. This work provides a great potential to extend the applications of aluminum alloys especially in marine engineering fields.

Supervised Learning of Two-Layer Perceptron under the Existence of External Noise — Learning Curve of Boolean Functions of Two Variables in Tree-Like Architecture —

We investigate the supervised batch learning of Boolean functions expressed by a two-layer perceptron with a tree-like structure. We adopt continuous weights (spherical model) and the Gibbs algorithm. We study the Parity and And machines and two types of noise, input and output noise, together with the noiseless case. We assume that only the teacher suffers from noise. By using the replica method, we derive the saddle point equations for order parameters under the replica symmetric (RS) ansatz. We study the critical value αC of the loading rate α above which the learning phase exists for cases with and without noise. We find that αC is nonzero for the Parity machine, while it is zero for the And machine. We derive the exponents \(\bar{\beta}\) of order parameters expressed as \((\alpha - \alpha _{\text{C}})^{\bar{\beta}}\) when α is near to αC. Furthermore, in the Parity machine, when noise exists, we find a spin glass solution, in which the overlap between the teacher and student vectors is zero but that...

Improvement of Pest Detection Using Histogram Adjustment Method and Gabor Wavelet

Plant diseases reduce production and cause huge economic losses in the agricultural section. At present, plant diseases are diagnosed by specialists and with the naked eyes. Presenting a simple, fast, cheap and accurate way for the diagnosis of plant diseases is necessary. In this paper, by using image processing and machine vision techniques, four methods for the diagnosis and classification of the diseases of corn leaf are presented. In the first method, the affected parts were separated from the healthy parts using the histogram adjustment; subsequently, a two-layer Perceptron Neural Network was used to categorize the final results and diagnose the disease type. The results indicate that Neural Network with an average of 65.15% is able to correctly diagnose the disease of the corn leaf. In the second method, different types of Laplacian filters, Canny and Sobel were applied on the leaves; after the separation of the affected parts, the classification and diagnosis phase were implemented. The results revealed that the algorithm with the accuracy of 67.94%, can correctly diagnose the disease. In the third method, using the analysis method of principal components, data dimension was reduced, and then was sent to the Support Vector Machine classifier for the diagnosis of the disease. This algorithm is able to correctly diagnose the disease with an accuracy of 75.28%. Furthermore, the algorithm is able to diagnose Class 5 and 4 diseases in a more accurate way. Finally, in the fourth method, a combination of Gabor filter and visual features was used in order to diagnose the type of disease. In this method, the proposed algorithm can correctly diagnose the disease with the accuracy of 90.04%.

Controlador neuronal para el seguimiento de trayectorias en un péndulo de rueda inercial

In this paper, the problem of trajectory tracking control in an inertia wheel pendulum is studied. Results are presented in a constructive form. First, a model-based controller is obtained by using the output feedback linearization technique. Then, the controller is redesigned by incorporating a neural network with the aim of avoiding the exact parameters knowledge of the inertia wheel pendulum, obtaining a robust control scheme. A two-layer perceptron is used, whose output weights are updated in real-time using an adaption law derived from the analysis of convergence of the closed-loop system solutions. Barbalat's lemma is used to conclude that the pendulum tracking error trajectory converges to zero. Numerical simulations and real-time experiments are presented, which confirm the theoretical results.

Evaluation HSE of a LSF system subject to near- and far-field earthquakes

The Lightweight Steel Framing (LSF) system has been proposed as an economical and earthquake resistant system. Due to the lightweight nature of LSF structures, the seismic performance of middle-rise buildings has been improved. Currently, various numerical-analytical methods have been proposed for seismic assessment of conventional structures. Providing a perfect seismic Health Safety Environment (HSE) index has always been regarded as one of the analytical passive points. In this research, an LSF building was selected as a case study for Finite Element (FE) modelling, in which non-linear time-history analyses were undertaken. Material properties were defined according to the performed experimental studies. A novel approach was presented for the seismic HSE index of LSF systems using the simultaneous incorporation of the non-linear analysis results and the correction coefficient describing the seismic geotechnical effects. The presented seismic HSE index accurately demonstrates the seismic performance of the LSF structures. Additionally, a two-layer perceptron Artificial Neural Network (ANN) was trained using the results of the FE model, and a non-linear relationship was obtained to predict the seismic damage index. Finally, the proposed seismic HSE index was validated using statistical analyses, indicating that the proposed method does not yield a significant difference compared to the ANN results.

System Identification in Difficult Operating Conditions Using Artificial Neural Networks

To investigate an ability of system identification in difficult operating conditions. A simulation based experiment was performed on a simple second order system with white noise signal superimposed to the output signal. Interferences are added to the output signal in order to simulate difficult operating conditions present in a real system environment. Based on system simulation measurements, the system was identified using conventional method with least squares estimate and an alternative method, a multi-layer perceptron (MLP) network. Graphical evaluation of simulation results showed that MLP network produced better results than conventional model, with significantly better results in case of interferences in the output signal. To model dynamic system, a simple two-layer perceptron network with external dynamic members was trained in Matlab using Levenberg-Marquardt algorithm.

Optimal Training Parameters and Hidden Layer Neuron Number of Two-Layer Perceptron for Generalised Scaled Object Classification Problem

The research is focused on optimising two-layer perceptron for generalised scaled object classification problem. The optimisation criterion is minimisation of inaccuracy. The inaccuracy depends on training parameters and hidden layer neuron number. After its statistics is accumulated, minimisation is executed by a numerical search. Perceptron is optimised additionally by extra training. As it is done, the classification error percentage does not exceed 3 % in case of the worst scale distortion.

A FRAMEWORK FOR CLASSIFIER SINGLE TRAINING PARAMETER OPTIMIZATION ON TRAINING TWO-LAYER PERCEPTRON IN A PROBLEM OF TURNED 60-BY-80-IMAGES CLASSIFICATION

A 13-itemed scenario framework for classifier single training parameter optimization is developed. Formally, the problem is to find global extremum (mostly, minimum) of function as a classifier output parameter against its single training parameter. Linking the scenario theory to praxis, the classifier type has been decided on two-layer perceptron. Its input objects are monochrome images of a medium format, having a few thousands independent features. Within the framework, the programming environment has been decided on MATLAB, having powerful Neural Network Toolbox. Keeping in mind the stochasticity of the being minimized function, there is defined statistical e-stability of its evaluation by a finite set of data. These data are mined in batch testings of the trained classifier. For exemplification of the scenario framework, there is optimized pixel-to-turn standard deviations ratio for training two-layer perceptron in classifying monochrome 60-by-80-images of the enlarged 26 English alphabet capital letters. The goal is to find a pixel-to-turn standard deviations ratio for the training process in order to ensure minimum of classification error percentage. The optimization relative gain is about a third. The developed framework can be applied also for classifier multivariable optimization, wherein it instructs which item operations shall regard the corresponding multiplicity of variables.

The Simulation of Microbial Enhanced Oil Recovery by Using a Two-layer Perceptron Neural Network

The authors simulated a reservoir by using two-layer perceptron. Indeed a model was developed to simulate the increase in oil recovery caused by bacteria injection into an oil reservoir. This model was affected by reservoir temperature and amount of water injected into the reservoir for enhancing oil recovery. Comparing experimental and simulation results and also the erratic trend of data show that the neural networks have modeled this system properly. Considering the effects of nonlinear factors and their erratic and unknown impacts on recovered oil, the perceptron neural network can develop a proper model for oil recovery factor in various conditions. The neural networks have not been applied in modeling of microbial enhanced oil recovery since now. Finally, we are going to design a controller for the neural network. This controller is designed for the case where output of the network is oil recovery factor. For this purpose, the network is designed as a one-layer network in which just one output matches each time. In this case, a one-layer network will have acceptable results.

A Non-Linear Blind Source Separation Method Based on Perceptron Structure and Conjugate Gradient Algorithm

The linear mixing model has been considered previously in most of the researches which are devoted to the blind source separation (BSS) problem. In practice, a more realistic BSS mixing model should be the non-linear one. In this paper, we propose a non-linear BSS method, in which a two-layer perceptron network is employed as the separating system to separate sources from observed non-linear mixture signals. The learning rules for the parameters of the separating system are derived based on the minimum mutual information criterion with conjugate gradient algorithm. Instead of choosing a proper non-linear functions empirically, the adaptive kernel density estimation is used in order to estimate the probability density functions and their derivatives of the separated signals. As a result, the score function of the perceptron's outputs can be estimated directly. Simulations show good performance of the proposed non-linear BSS algorithm.

Vector-Matrix Multiply and Winner-Take-All as an Analog Classifier

The vector-matrix multiply and winner-take-all structure is presented as a general-purpose, low-power, compact, programmable classifier architecture that is capable of greater computation than a one-layer neural network, and equivalent to a two-layer perceptron. The classifier generates event outputs and is suitable for integration with event-driven systems. The main sources of mismatch, temperature dependence, and methods for compensation are discussed. We present measured data from simple linear and nonlinear classifier structures on a 0.35-μm chip and analyze the power and computing efficiency for scaled structures.

Seismic strain analysis of buried pipelines in a fault zone using hybrid FEM-ANN approach

This study was concerned on the application of a hybrid approach for analyzing the buried pipelines deformations subjected to earthquakes. Nonlinear time-history analysis of Finite Element (FE) model of buried pipelines, which was modeled using laboratory data, has been performed via selected earthquakes. In order to verify the FE model with experiments, a statistical test was done which demonstrated a good conformity. Then, the FE model was developed and the optimum intersection angle of pipeline and fault was obtained via genetic algorithm. Transient seismic strain of buried pipeline in the optimum intersection angle of pipeline and fault was investigated considering the pipes diameter, the distance of pipes from fault, the soil friction angles and seismic response duration of buried pipelines. Also, a two-layer perceptron Artificial Neural Network (ANN) was trained using results of FE model, and a nonlinear relationship was obtained to predict the bending strain of buried pipelines based on the pipes diameter, intersection angles of the pipelines and fault, the soil friction angles, distance of pipes from the fault, and seismic response duration; whereas it contains a wide range of initial input data without any requirement to laboratory measurements.

Optimization of multilayer perspetronu systems of information protection

The Article is dedicated to development of the methods to optimization of the structure laminated perceptron intended for use in computer system of protection to information for the reason monitoring and management their parameter. Proved practicability of the use two-layer perceptron. Accounting expressions is Received for determination amount neuron in hidden layer.

Study on Optical Performance Monitoring for Fiber-Optical Link Utilizing Chaos Theory

Optical-link performance monitoring (OLPM) plays a crucial role in ensuring the link performance, the channel assignment and the network reconfiguration in the high-speed and WDM optical networks. An OLPM scheme based on the chaos theory and the pattern recognition is proposed in this paper. The identification results of nonlinear regimes of 4/8/16-channel WDM links with 10 Gb/s are demonstrated through utilizing the maximum Lyapunov exponent, the recurrence plot and the two-layer perceptron.

Grounding the Meanings in Sensorimotor Behavior using Reinforcement Learning.

The recent outburst of interest in cognitive developmental robotics is fueled by the ambition to propose ecologically plausible mechanisms of how, among other things, a learning agent/robot could ground linguistic meanings in its sensorimotor behavior. Along this stream, we propose a model that allows the simulated iCub robot to learn the meanings of actions (point, touch, and push) oriented toward objects in robot’s peripersonal space. In our experiments, the iCub learns to execute motor actions and comment on them. Architecturally, the model is composed of three neural-network-based modules that are trained in different ways. The first module, a two-layer perceptron, is trained by back-propagation to attend to the target position in the visual scene, given the low-level visual information and the feature-based target information. The second module, having the form of an actor-critic architecture, is the most distinguishing part of our model, and is trained by a continuous version of reinforcement learning to execute actions as sequences, based on a linguistic command. The third module, an echo-state network, is trained to provide the linguistic description of the executed actions. The trained model generalizes well in case of novel action-target combinations with randomized initial arm positions. It can also promptly adapt its behavior if the action/target suddenly changes during motor execution.

Neural input space mapping optimization based on nonlinear two-layer perceptrons with optimized nonlinearity

A neural space mapping optimization algorithm based on nonlinear two layer perceptrons (2LP) is described in this article. This work is an improved version of the Neural Space-Mapping (NSM) algorithm that uses three layer perceptrons (3LP) to implement a nonlinear input mapping function at each iteration. The new version uses a nonlinear 2LP whose nonlinearity is automatically regulated with classical optimization algorithms. Additionally, the new algorithm uses a different optimization method to train the SM-based neuromodel and a more efficient manner to predict the next iterate. With these improvements, we obtain a more efficient and faster algorithm. To verify the algorithm performance, we design some synthetic circuits, as well as a stopband microstrip filter with quarter-wave resonant opens stubs, a bandpass microstrip filter, and a microstrip notch filter with mitered bends. The last three cases use commercially available full-wave electromagnetic simulators. A rigorous comparison is made with the original NSM algorithm, showing the performance improvement achieved by our proposed new formulation. © 2010 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2010.

A class possibility based kernel to increase classification accuracy for small data sets using support vector machines

Appropriate choice of kernels is the most important task when using kernel-based learning methods such as support vector machines. The current widely used kernels (such as polynomial kernel, Gaussian kernel, two-layer perceptron kernel, and so on) are all functional kernels for general purposes. Currently, there is no kernel proposed in a data-driven way. This paper proposes a new kernel generating method dependent on classifying related properties of the data structure itself. The new kernel concentrates on the similarity of paired data in classes, where the calculation of similarity is based on fuzzy theories. The experimental results with four medical data sets show that the proposed kernel has superior classification performance than polynomial and Gaussian kernels.

Syntactic systematicity in sentence processing with a recurrent self-organizing network

As potential candidates for explaining human cognition, connectionist models of sentence processing must demonstrate their ability to behave systematically, generalizing from a small training set. It has recently been shown that simple recurrent networks and, to a greater extent, echo-state networks possess some ability to generalize in artificial language learning tasks. We investigate this capacity for a recently introduced model that consists of separately trained modules: a recursive self-organizing module for learning temporal context representations and a feedforward two-layer perceptron module for next-word prediction. We show that the performance of this architecture is comparable with echo-state networks. Taken together, these results weaken the criticism of connectionist approaches, showing that various general recursive connectionist architectures share the potential of behaving systematically.

Identification of a Continuous Stirring Tank Reactor Using Two-layer Perceptron Neural Network Model

This paper investigates the identification of a continuous stirring tank reactor using a two-layer multi-layer perceptron neural network. It is shown that the neural network based state-space innovations form model clearly outperforms the equivalent linear model in simulation performance and correlation functions of prediction errors and thus, it is possible to obtain hopeful results for this system with neural network based state-space innovations form model. It is by no means claimed that the “optimal” solution is found, however, it is shown that the proposed neural network based model has not only significantly performed better in simulation but also in correlation characteristics than the equivalent linear model in the sense that the correlation functions tend to lie more closely within the confidence regions. This also means that most of the information has been extracted from the training set and that the model approximates the given system well.

Prediction of a Lorenz chaotic attractor using two-layer perceptron neural network

This paper investigates the prediction of a Lorenz chaotic attractor having relatively high values of Lypunov's exponents. The characteristic of this time series is its rich chaotic behavior. For such dynamic reconstruction problem, regularized radial basis function (RBF) neural network (NN) models have been widely employed in the literature. However, author recommends using a two-layer multi-layer perceptron (MLP) NN-based recurrent model. When none of the available linear models have been able to learn the dynamics of this attractor, it is shown that the proposed NN-based auto regressive (AR) and auto regressive moving average (ARMA) models with regularization have not only learned the true trajectory of this attractor, but also performed much better in multi-step-ahead predictions. However, equivalent linear models seem to fail miserably in learning the dynamics of the time series, despite the low values of Akaike's final prediction error (FPE) estimate. Author proposes to employ the recurrent NN-based ARMA model with regularization which clearly outperforms all other models and thus, it is possible to obtain good results for prediction and reconstruction of the dynamics of the chaotic time series with NN-based models.