Neural Smithing

Abstract

Artificial neural networks are nonlinear mapping systems whose structure is loosely based on principles observed in the nervous systems of humans and animals. The basic idea is that massive systems of simple units linked together in appropriate ways can generate many complex and interesting behaviors. This book focuses on the subset of feedforward artificial neural networks called multilayer perceptrons (MLP). These are the mostly widely used neural networks, with applications as diverse as finance (forecasting), manufacturing (process control), and science (speech and image recognition). This book presents an extensive and practical overview of almost every aspect of MLP methodology, progressing from an initial discussion of what MLPs are and how they might be used to an in-depth examination of technical factors affecting performance. The book can be used as a tool kit by readers interested in applying networks to specific problems, yet it also presents theory and references outlining the last ten years of MLP research. Bradford Books imprint

Predictive modeling of surface roughness and tool wear in hard turning using regression and neural networks

“…Methods such as Bayesian regularization, early stopping, etc. are commonly used to improve the generalization in neural networks [37]. In this study, the Levenberg-Marquardt method is used together with Bayesian regularization in training neural networks in order to obtain neural networks with good generalization capability.…”

“…In this study, the Levenberg-Marquardt method is used together with Bayesian regularization in training neural networks in order to obtain neural networks with good generalization capability. The details of Levenberg-Marquardt algorithm can be found in [37].…”

Comparing evolutionary hybrid systems for design and optimization of multilayer perceptron structure along training parameters

“…The representation ability of a neural network depends on the number of layers, on the number of neurons per layer and on the connectivity between layers. It was demonstrated that a network with two hidden layers can solve any pattern classification problem [65,6,94]. On the other hand, several authors have proved that any function approximation problem can be solved by using one hidden layer [60,38,37,23].…”

“…Now, however, the dilemma of overfitting arises: small networks generalize well, but are slow at learning, whereas big networks learn fast (needing fewer training epochs to obtain similar precision), but generalize badly [5,4]. The way to obtain good generalization ability is to use the smallest network that can learn the input data efficiently [93,94,103]. For this reason, this operator is combined with the following one.…”

“…Generalization ability may vary during the training process, to the extent that at a given point, the generalization error reaches a minimum, after which it starts to grow [94] when the network learns some characteristics that are not representative of the training set. Thus, early detection of overfitting occurring during supervised training is important in order to prevent the error from increasing.…”