INTERVAL POLYNOMIAL REGRESSION BY USE OF A NEURAL-NETWORK FOR MINIMUMZONE PROBLEMS

Determining the parametrization of the curve is a fundamental problem in approximation and interpolation. The goal of this paper is to devel op an accurate and robust algorithm for the minimum zone problems. In this paper, we use an interval bias adaptive linear neural network str ucture together with an appropriate cost function and the least mean s quares learning algorithm to carry out the interval regression analysi s. Through appropriate choice of the output function of the input neur on, the interval polynomial regression use of a neural network (IPRNN) method developed in this paper is applicable to many problems (interv al algebraic polynomial approximation, evaluation of straightness, rou ndness and ellipticity and so on). Generally, these problems have comp licated constraints and the LSQ method cannot be used.