Neural, Parallel & Scientific Computations, Volume 1, 1993.

Neural, Parallel & Scientific Computations

Volume 1, 1993

CONTENTS

Editorial	1
Sergey K. Aityan and David L. Barrow Paradigm, logical performance, and training of recurrent refractory neural networks	3
D. J. Evans and W. S. Yousif The parallel strides reduction algorithms for solving tridiagonal linear systems	29
Li Yuanxiang,Kang Lishan and Wu Zhijian Massively parallel algorithm I: a new class of lattice gas methods	43
V. Eigel-Danielson and M. F. Augusteijn Neural network pruning and its effect on generalization, some experimental results	59
Pong P. Chu Neural network for solving optimization problems with linear equality constraints	71
I. Grabec Optimization of kernel-type density estimator by the principle maximal self-consistency	83
C. H. Chen and G. H. You Class sensitive neural networks	93
P.R. Parthasarathy and V. Vijayalakshmi Transient solution of state dependent queues-a continued fraction approach	97
A. Bulsari and H. Saxen Control of a biochemical process using feed-forward neural networks	109
Tep Sastri and Gi-Nam Wang An adaptive model identification method	117
H. Adeli and J. Zhang An improved perception learning algorithm	141
C. Y. Chan Computation of the critical domain for quenching in an elliptic plate	153
Yuri P. Boglaev Logic programming in the numerical algorithm design for parallel computations	163
Yasuo Matsuyama Competitive learning parallel agents: application to travelling salesperson problem	181
Thiab R. Taha and Peiqing Jiang Parallel algorithms for solving banded Toeplitz linear systems	199
H. Byun, E. N. Houstis and S. K. Kortesis A workload partitioning strategy for PDE computations by a generalized neural network	209
P. Saratchandran Effect of hidden layers on generalization properties of feedforward neural networks	227
Sameh E. Rehan and Mohamed I. Elmasry Modular switched-resistor ANN chip for character recognition using novel parallel VLSI architecture	241
Dusan Caf and David J. Evans A new UL factorisation of banded toeplitz and Hankel matrices	263
S.E. Mentzelopoulou, R. Hormis, G.E. Antoniou, & C.N. Manikopoulos On the factorization of 2-D polynomials via neural networks	271
C. Lujuan, I. Macleod & E. V. Krishnamurthy Suffix parallel computation and operator splitting for fast solution of PDE's	287
H. Byun, S.K. Kortesis, E.N. Houstis & E.A. Vavalis A virtual parallel environment for implementing neural network computations on parallel machines	301
V. Kreinovich and O. Sirisaengtaksin 3-layer neural networks are universal approximations for functions and for control strategies	325
P. Raveendran, S. Omatu, & A.B. Wan Neural network approach to minimize intraclass invariance of moment invariant features for pattern classification	347
W.P. Mounfield, Jr & L. T. Grujic Natural tracking control for exponential tracking: lateral high-gain pi control of an aircraft system with state-space description	357
Gary Chung, N. Medhin, & M. Sambandham A neural network approach for solving certain control problem	371
Xun Liang Method of digging tunnels into the error hypersurface	381
Y. Jin, A. G. Pipe, A. Winfield Neural networks for prediction and control of discrete systems	395
Jun He and Lishan Kang Parallel multigrid algorithms based on wavelet multiresolution approximations	421
Sergey K. Aityan Correlation-recall operation in encoding fuzzy associative memories	431
Yi Jiang and Alden H. Wright O(k) parallel algorithm for approximate string matching	443
E. Andrew Boyd Solving integer programs with Fenchel cutting planes and preprocessing	453
D.G. Lainiotis, N.D. Assimakis, and S.K. Katsikas Fast and stable algorithm for computing the principal square root of a complex matrix	467
D.G. Lainiotis, N.D. Assimakis, and S.K. Katsikas Time series prediction based on fractal theory and nonlinear estimation using a network with gaussian kernel functions	477
Contents of Volume 1	501
Authors' Index	505