Fundamenta Informaticae - Volume 37, issue 1-2

Authors: Pal, Sankar K.

Article Type: Other

DOI: 10.3233/FI-1999-371210

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. v-vii, 1999

Authors: Dubois, Didier | Berre, Daniel Le | Prade, Henri | Sabbadin, Régis

Article Type: Research Article

Abstract: This paper describes a logical machinery for computing decisions, where the available knowledge on the state of the world is described by a possibilistic propositional logic base (i.e., a collection of logical statements associated with qualitative certainty levels), and where the preferences of the user are also described by another possibilistic logic base whose formula weights are interpreted in terms of priorities. Two attitudes are allowed for the decision maker: a pessimistic risk-averse one and an optimistic one. The computed decisions are in agreement with a qualitative counterpart to the classical theory of expected utility, recently developed by three of …the authors. A link is established between this logical view of qualitative decision making and an ATMS-based computation procedure. Efficient algorithms for computing pessimistic and optimistic optimal decisions are finally given in this logical setting (using some previous work of the fourth author). Show more

Keywords: qualitative decision, possibilistic logic, possibility theory, ATMS

DOI: 10.3233/FI-1999-371201

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 1-30, 1999

Authors: Sinha, Sitabhra

Article Type: Research Article

Abstract: The paper examines the discrete-time dynamics of neuron models (of excitatory and inhibitory types) with piecewise linear activation functions, which are connected in a network. The properties of a pair of neurons (one excitatory and the other inhibitory) connected with each other, is studied in detail. Even such a simple system shows a rich variety of behavior, including high-period oscillations and chaos. Border-collision bifurcations and multifractal fragmentation of the phase space is also observed for a range of parameter values. Extension of the model to a larger number of neurons is suggested under certain restrictive assumptions, which makes the resultant …network dynamics effectively one-dimensional. Possible applications of the network for information processing are outlined. These include using the network for auto-association, pattern classification, nonlinear function approximation and periodic sequence generation. Show more

Keywords: excitatory-inhibitory neural networks, chaos, nonlinear dynamics

DOI: 10.3233/FI-1999-371202

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 31-50, 1999

Authors: Kubota, Naoyuki | Fukuda, Toshio

Article Type: Research Article

Abstract: This paper deals with an ecological model on planar gird of a genetic algorithm based on virus theory of evolution (E-VE-GA). In the E-VE-GA, each individual is placed on a planar grid and genetic operators are performed between neighborhoods. The E-VE-GA can self-adaptively change searching ratio between global and local searches. The main operator of the E-VE-GA is reverse transcription and incorporation transmitting local genetic information. The convergence of the E-VE-GA depends on the frequency and localization of the virus infection. In this paper, we apply the E-VE-GA to traveling salesman problems and discuss the coevolution of host and virus …populations through the numerical simulation. Show more

Keywords: evolutionary computation, optimization, virus theory of evolution, traveling salesman problem

DOI: 10.3233/FI-1999-371203

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 51-70, 1999

Authors: Tanaka, Hideo | Lee, Haekwan

Article Type: Research Article

Abstract: This paper proposes interval regression analysis with polynomials. For data sets with crisp inputs and interval outputs, three estimation models called as an upper, a lower, and a possibility estimation models can be formulated from the concepts of the possibility and necessity measures. Always there exists an upper and a possibility estimation model when a linear system with interval coefficients is considered, but it is not assured to attain a solution for a lower estimation model in an interval linear system. If we can not obtain the lower estimation model, it might be caused by adopting a model not fitting …to the given data. Thus we consider polynomials to find a regression model which fits well to the given observations. The possibility model is used to check the existence of the lower model. If we can find a proper lower model, the estimated upper and lower models deserve more credit than the previous models in the former studies. We also introduce the measure of fitness to gauge the degree of approximation of the obtained models to the given data. The upper and lower estimation models in interval regression analysis can be considered as the upper and lower approximations in rough sets. The similarity between the interval estimation models and the rough sets concept is also discussed. In order to illustrate our approach, numerical examples are shown. Show more

Keywords: interval regression analysis, possibility and necessity measures, upper, lower, and possibility estimation models, rough sets, upper and lower approximations

DOI: 10.3233/FI-1999-371204

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 71-87, 1999

Authors: Friedman, Menahem | Ming, Ma | Kandel, Abraham

Article Type: Research Article

Abstract: Using the embedding method, the fuzzy integral is represented as a parametric Riemann integral. An algorithm which approximates this integral uniformly, is incorporated to design a soft computing tool for solving a fuzzy Fredholm integral equation of the second kind with arbitrary kernel, using a uniformly convergent iterative procedure.

Keywords: fuzzy number, Riemann integral, fuzzy integral, numerical integration, soft computing

DOI: 10.3233/FI-1999-371205

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 89-99, 1999

Authors: Gesù, Vito Di

Article Type: Research Article

Abstract: The term soft-computing has been introduced by Zadeh in 1994. Soft-computing provides an appropriate paradigm to program malleable and smooth concepts. For example, it can be used to introduce flexibility in artificial systems to improve their Intelligent Quotient. The aim of this paper is to describe the applicability of soft-computing to artificial vision problems. Good performance of this approach is assured by the fact that digital images are examples of fuzzy entities, where shapes are not always describable by exact equations and their approximation can be very complex.

Keywords: computer vision, soft computing, fuzzy sets, mathematical morphology, image segmentation

DOI: 10.3233/FI-1999-371206

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 101-119, 1999

Authors: Sastry, K.K.N. | Behera, L. | Nagrath, I.J.

Article Type: Research Article

Abstract: This paper presents an unconventional approach to adaptive fuzzy logic controller (FLC) design wherein a new evolution strategy, Differential Evolution (DE) is used in the simultaneous design of membership functions and rule sets for fuzzy logic controllers. Differential Evolution is an exceptionally simple, fast, and robust population based search algorithm that is able to locate near-optimal solutions to difficult problems. This technique, which is similar to genetic algorithms, has been applied to the control of pH, which is a requirement in many chemical industries. Control of pH poses a difficult problem because of inherent nonlinearities and frequently changing process dynamics. …This technique has been successfully implemented on a laboratory scale pH plant setup. The results have been compared with a simple GA based adaptive FLC where we have incorporated a search space smoothing function for achieving faster convergence and for ascertaining a global optimum. Results indicate that FLC’s augmented with DE’s offer a powerful alternative to GA based FLC’s. Results also show that the search space smoothing function helps in faster convergence of a GA. Show more

Keywords: Adaptive Control, Differential Evolution, Fuzzy Logic Control, Genetic Algorithms, Nonlinear Control, pH Control, Process Control, Smoothing Function

DOI: 10.3233/FI-1999-371207

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 121-136, 1999

Authors: Karayiannis, Nicolaos B.

Article Type: Research Article

Abstract: This paper proposes a framework for developing a broad variety of soft clustering and learning vector quantization (LVQ) algorithms based on gradient descent minimization of a reformulation function. According to the proposed axiomatic approach to learning vector quantization, the development of specific algorithms reduces to the selection of a generator function. A linear generator function leads to the fuzzy c-means (FCM) and fuzzy LVQ (FLVQ) algorithms while an exponential generator function leads to entropy constrained fuzzy clustering (ECFC) and entropy constrained LVQ (ECLVQ) algorithms. The reformulation of clustering and LVQ algorithms is also extended to supervised learning models through an …axiomatic approach proposed for reformulating radial basis function (RBF) neural networks. This approach results in a broad variety of admissible RBF models, while the form of the radial basis functions is determined by a generator function. This paper shows that gradient descent learning makes reformulated RBF neural networks an attractive alternative to conventional feed-forward neural networks. Show more

Keywords: fuzzy clustering, learning vector quantization, reformulation, generator function, radial basis neural networks, function approximation, gradient descent learning

DOI: 10.3233/FI-1999-371208

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 137-175, 1999

Authors: Bandyopadhyay, Sanghamitra | Pal, Sankar K.

Article Type: Research Article

Abstract: An analogy between a genetic algorithm based pattern classification scheme (where hyperplanes are used to approximate the class boundaries through searching) and multilayer perceptron (MLP) based classifier is established. Based on this, a method for determining the MLP architecture automatically is described. It is shown that the architecture would need atmost two hidden layers, the neurons of which are responsible for generating hyperplanes and regions. The neurons in the second hidden and output layers perform the AND & OR functions respectively. The methodology also includes a post processing step which automatically removes any redundant neuron in the hidden/output layer. An …extensive comparative study of the performance of the MLP, thus derived using the proposed method, with those of several other conventional MLPs is presented for different data sets. Show more

Keywords: hyperplane fitting, boundary approximation, hard limiting neuron, network architecture design, variable string length genetic algorithm

DOI: 10.3233/FI-1999-371209

Citation: Fundamenta Informaticae, vol. 37, no. 1-2, pp. 177-199, 1999