Multiple optimal learning factors for the multi-layer perceptron

Malalur, Sanjeev S.; Manry, M.T.; Jesudhas, Praveen

doi:10.1016/j.neucom.2014.08.043

Cited by 16 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The size of SHL in 2LP for the problem of classifying SDM6080I can be preset up to 240 neurons. It is an adequate SHL size for GT 5, and such size ensures reasonable speed and accuracy [3], [6], [10], [11], [13]. Then 2LP configuration is 4800-240-26 (input layer neuron number -SHL neuron numberoutput layer neuron number), and this 4800-240-26 perceptron (4800-240-26-P) is initialised on MATLAB Neural Network Toolbox simply with function "feedforwardnet" or "newff".…”

Section: Number Of Neurons In Perceptron Hidden Layersmentioning

confidence: 99%

See 1 more Smart Citation

Optimal Pixel-to-Shift Standard Deviation Ratio for Training 2-Layer Perceptron on Shifted 60 × 80 Images with Pixel Distortion in Classifying Shifting-Distorted Objects

Romanuke

2016

Applied Computer Systems

View full text Add to dashboard Cite

An optimization problem of classifying shifting-distorted objects is studied. The classifier is 2-layer perceptron, and the object model is monochrome 60 × 80 image. Based on the fact that previously the perceptron has successfully been attempted to classify shifted objects with a pixel-to-shift standard deviation ratio for training, the ratio is optimized. The optimization criterion is minimization of classification error percentage. A classifier trained under the found optimal ratio is optimized additionally. Then it effectively classifies shifting-distorted images, erring only in one case from eight takings at the maximal shift distortion. On average, classification error percentage appears less than 2.5 %. Thus, the optimized 2-layer perceptron outruns much slower neocognitron. And the found optimal ratio shall be nearly the same for other object classification problems, when the number of object features varies about 4800, and the number of classes is between two and three tens.

show abstract

Section: Number Of Neurons In Perceptron Hidden Layersmentioning

confidence: 99%

“…Pixel-to-shift standard deviation ratio (PSSDR) in (3) was the constant for the given 2LP and SDO classification problem. It was stated that the better PSSDR (3) was adjusted [12], [13] the lower CEP was going to be performed by 2LP in classifying SDO, modelled here as SDM6080I. Under ; rr is to be evaluated rather than approximated.…”

mentioning

confidence: 99%

Optimal Pixel-to-Shift Standard Deviation Ratio for Training 2-Layer Perceptron on Shifted 60 × 80 Images with Pixel Distortion in Classifying Shifting-Distorted Objects

Romanuke

2016

Applied Computer Systems

View full text Add to dashboard Cite

show abstract

“…Los pesos de entra-Determinación del estado de maduración de frutos de feijoa mediante un sistema de visión por computador utilizando información de color da conectan con cada una de las neuronas de la capa oculta y los pesos de salida conectan con la activación de cada una de las salidas. Los datos de entrenamiento son descritos por el conjunto de elementos en la capa de entrada lo cual conforma un vector de entrada N-dimensional, y los vectores de salida M-dimensionales en función de las clases o conjuntos deseados (Malaur, Manry, & Jesudhas, 2015). Ellas se han utilizado ampliamente en aplicaciones de regresión y clasificación en áreas como: estimación de parámetros, análisis de documentos y reconocimiento, finanzas, fabricación y minería de datos.…”

Section: Clasificadoresunclassified

Determinación del estado de maduración de frutos de feijoa mediante un sistema de visión por computador utilizando información de color

Bonilla-González

Prieto

2016

Rev. investig. desarro. innov

View full text Add to dashboard Cite

Determinar el estado de madurez de productos agrícolas, generalmente depende de un análisis realizado por expertos humanos. La decisión final sobre el estado de madurez donde se encuentra el producto, resulta de correlacionar algunas de sus propiedades físicas con características químicas e internas del fruto. La necesidad de preservar la integridad del fruto en dicho análisis, hace necesario la implementación de tecnologías que emitan un juicio sobre el estado del mismo, sin necesidad de destruirlo. El uso del índice de color como propiedad física, contribuye a la solución de este problema. En este documento, se presenta un sistema de visión por computador para clasificar en tres estados de madurez un fruto exótico específico, feijoa -Acca Sellowiana-. Los resultados obtenidos a partir de la clasificación, utilizando diferentes clasificadores, permiten obtener una respuesta superior al 90%, para 156 imágenes de frutos de feijoa utilizadas en el estudio.

show abstract

“…It is a distributed paradigm, which seeks to simulate the collective behavior of unsophisticated individuals interacting locally with their environment to efficiently identify optimum solutions in complex search spaces. There are many related works of research [22][23][24][25] which show that the BBO algorithm is a type of evolutionary algorithm which can offer a specific evolutionary mechanism for each individual in a population. This mechanism makes the BBO algorithm more successful and robust on nonuniform training procedures than gradient-based algorithms.…”

Section: Introductionmentioning

confidence: 99%

Developing a Novel Hybrid Biogeography-Based Optimization Algorithm for Multilayer Perceptron Training under Big Data Challenge

Chen

et al. 2018

Scientific Programming

View full text Add to dashboard Cite

A Multilayer Perceptron (MLP) is a feedforward neural network model consisting of one or more hidden layers between the input and output layers. MLPs have been successfully applied to solve a wide range of problems in the fields of neuroscience, computational linguistics, and parallel distributed processing. While MLPs are highly successful in solving problems which are not linearly separable, two of the biggest challenges in their development and application are the local-minima problem and the problem of slow convergence under big data challenge. In order to tackle these problems, this study proposes a Hybrid Chaotic Biogeography-Based Optimization (HCBBO) algorithm for training MLPs for big data analysis and processing. Four benchmark datasets are employed to investigate the effectiveness of HCBBO in training MLPs. The accuracy of the results and the convergence of HCBBO are compared to three well-known heuristic algorithms: (a) Biogeography-Based Optimization (BBO), (b) Particle Swarm Optimization (PSO), and (c) Genetic Algorithms (GA). The experimental results show that training MLPs by using HCBBO is better than the other three heuristic learning approaches for big data processing.

show abstract

Multiple optimal learning factors for the multi-layer perceptron

Cited by 16 publications

References 24 publications

Optimal Pixel-to-Shift Standard Deviation Ratio for Training 2-Layer Perceptron on Shifted 60 × 80 Images with Pixel Distortion in Classifying Shifting-Distorted Objects

Optimal Pixel-to-Shift Standard Deviation Ratio for Training 2-Layer Perceptron on Shifted 60 × 80 Images with Pixel Distortion in Classifying Shifting-Distorted Objects

Determinación del estado de maduración de frutos de feijoa mediante un sistema de visión por computador utilizando información de color

Developing a Novel Hybrid Biogeography-Based Optimization Algorithm for Multilayer Perceptron Training under Big Data Challenge

Contact Info

Product

Resources

About