Operador de Hardy-Littlewood reiterado y clases de Orlicz de funciones maximales.
Una nueva definición de aplicación difusa.
If X, Y are universes of discourse, a fuzzy mapping f: X --> Y is defined as a classical mapping f: X x [0,1] --> P(Y). Their basic properties are studied as well as their relations with the classical model of fuzzy mapping.
Max-min fuzzy neural networks for solving relational equations.
The Relational Equations approach is one of the most usual ones for describing (Fuzzy) Systems and in most cases, it is the final expression for other descriptions. This is why the identification of Relational Equations from a set of examples has received considerable atention in the specialized literature. This paper is devoted to this topic, more specifically to the topic of max-min neural networks for identification. Three methods of learning Fuzzy Systems are developed by combining the most...
Fuzzy numbers, definitions and properties.
Two different definitions of a Fuzzy number may be found in the literature. Both fulfill Goguen's Fuzzification Principle but are different in nature because of their different starting points. The first one was introduced by Zadeh and has well suited arithmetic and algebraic properties. The second one, introduced by Gantner, Steinlage and Warren, is a good and formal representation of the concept from a topological point of view. The objective of this paper is to analyze...
Neural methods for obtaining fuzzy rules.
In previous papers, we presented an empirical methodology based on Neural Networks for obtaining fuzzy rules which allow a system to be described, using a set of examples with the corresponding inputs and outputs. Now that the previous results have been completed, we present another procedure for obtaining fuzzy rules, also based on Neural Networks with Backpropagation, with no need to establish beforehand the labels or values of the variables that govern the system.
New aspects on extraction of fuzzy rules using neural networks.
In previous works, we have presented two methodologies to obtain fuzzy rules in order to describe the behaviour of a system. We have used Artificial Neural Netorks (ANN) with the Backpropagation algorithm, and a set of examples of the system. In this work, some modifications which allow to improve the results, by means of an adaptation or refinement of the variable labels in each rule, or the extraction of local rules using distributed ANN, are showed. An interesting application on the assignement...
A survey of methods to evaluate quantified sentences.
The evaluation of quantified sentences is used to solve several problems. Most of the methods proposed in the literature are not satisfactory because they do not verify some intuitive properties. In this paper we propose an extension of both possibilistic and probabilistic methods, based on the Sugeno and the Choquet fuzzy integrals respectively, for the evaluation of type II sentences, the most general kind of sentences. These methods verify good properties, and they are shown to be better than...
El problema del árbol minimal para grafos difusos.
Basándonos en algunas definiciones previas, se analiza el problema del árbol generador difuso. En primer lugar se trata su existencia y después se encuentra el árbol generador difuso de mínimo costo mediante una descomposición por α-cortes. El estudio se realiza para dos estructuras diferentes de costos.
Resolución por programación paramétrica del problema multiobjetivo lineal difuso.
En este artículo se propone una solución difusa al problema Multiobjetivo Lineal Difuso. Tal solución contiene, como valores particulares, las soluciones puntuales que otros autores han obtenido. El método que se emplea es independiente de las funciones de pertenencia que se consideren. El problema también se extiende al caso en que el conjunto de restricciones sea, junto con los objetivos, difuso.
Evolutionary training for Dynamical Recurrent Neural Networks: an application in finantial time series prediction.
Theoretical and experimental studies have shown that traditional training algorithms for Dynamical Recurrent Neural Networks may suffer of local optima solutions, due to the error propagation across the recurrence. In the last years, many researchers have put forward different approaches to solve this problem, most of them being based on heuristic procedures. In this paper, the training capabilities of evolutionary techniques are studied, for Dynamical Recurrent Neural Networks. The performance...
Page 1