Publicación: Optimización de Parámetros en Extreme Learning Machine mediante Algoritmos Evolutivos
Cargando...
Fecha
2023-06
Autores
Editor/a
Director/a
Tutor/a
Coordinador/a
Prologuista
Revisor/a
Ilustrador/a
Derechos de acceso
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Educación a Distancia (UNED)
Resumen
Extreme Learning Machine es un paradigma relativamente reciente. Específicamente, se trata de una técnica que utiliza redes neuronales en las que los pesos de entrada son generados aleatoriamente y los pesos de salida son obtenidos mediante la matriz inversa generalizada de Moore-Penrose de la matriz de salida de la capa oculta. El tiempo de entrenamiento de este algoritmo es mucho menor que el de las redes neuronales típicas y su rendimiento no se ve afectado por ello, de ahí que los últimos años hayan sido el centro de atención de numerosas investigaciones. Algunos de estos trabajos se benefician de este bajo coste computacional en el entrenamiento para aplicar algoritmos de computación evolutiva que sintonicen adecuadamente los hiperparámetros de este tipo de redes con el n mejorar su potencia predictiva o de reducir su complejidad. Este trabajo se enmarca en esta última línea de investigación, teniendo como principal objetivo el uso de un algoritmo de computación evolutiva, denominado Covariance Matrix Adaptation Evolution Strategy, para caracterizar la función de activación más adecuada para cada conjunto de datos utilizado, con el n último de mejorar (aumentando la potencia predictiva o reduciendo la complejidad) el modelo obtenido mediante Extreme Learning Machine.
Extreme Learning Machine is a relatively recent paradigm. Speci cally, it is a technique that uses neural networks in which the input weights are randomly generated and the output weights are obtained using the generalized Moore-Penrose inverse matrix of the hidden layer output matrix. The training time of this algorithm is much shorter than that of typical neural networks and its performance is not a ected by it, hence in recent years it has been the focus of much research. Some of these works take advantage of this low computational cost in training to apply evolutionary computation algorithms that adequately tune the hyperparameters of this type of networks to improve their predictive power or to reduce their complexity. This work is part of the latter line of research, having as main objective the use of an evolutionary computation algorithm, called Covariance Matrix Adaptation Evolution Strategy, to characterize the most appropriate activation function for each data set used, with the ultimate goal of improving (increasing the predictive power or reducing the complexity) the model obtained by Extreme Learning Machine
Extreme Learning Machine is a relatively recent paradigm. Speci cally, it is a technique that uses neural networks in which the input weights are randomly generated and the output weights are obtained using the generalized Moore-Penrose inverse matrix of the hidden layer output matrix. The training time of this algorithm is much shorter than that of typical neural networks and its performance is not a ected by it, hence in recent years it has been the focus of much research. Some of these works take advantage of this low computational cost in training to apply evolutionary computation algorithms that adequately tune the hyperparameters of this type of networks to improve their predictive power or to reduce their complexity. This work is part of the latter line of research, having as main objective the use of an evolutionary computation algorithm, called Covariance Matrix Adaptation Evolution Strategy, to characterize the most appropriate activation function for each data set used, with the ultimate goal of improving (increasing the predictive power or reducing the complexity) the model obtained by Extreme Learning Machine
Descripción
Categorías UNESCO
Palabras clave
Extreme Learning Machine, Evolutionary Extreme Learning Machine, Computación evolutiva, CMA-ES, función de activación, MNIST, Evolutionary computing, activation functions
Citación
Pinto Lozano, José Manuel, (2023) Optimización de Parámetros en Extreme Learning Machine mediante Algoritmos Evolutivos. Trabajo Fin de Máster. Universidad de Educación a Distancia (UNED)
Centro
Facultades y escuelas::E.T.S. de Ingeniería Informática
Departamento
Inteligencia Artificial