Bp. Dubey et al., NEURAL-NETWORK FITS TO NEUTRON-INDUCED REACTIONS USING WEIGHTED LEAST-MEAN-SQUARES, Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment, 397(2-3), 1997, pp. 426-439
A new error function based on weighted least-mean-square analysis is s
uggested for use in a bask-propagation algorithm in neural networks. T
he standard back-propagation algorithm has been revised to minimize th
e sum of the weighted residual error squared. This method has been app
lied to obtain the cross-section For 14.6 MeV neutron induced (n, p) a
nd (n, alpha) reactions in the mass region 40 < A < 240. In comparison
to the normal back-propagation algorithm based on simple least mean s
quare error estimation? the proposed method has better estimation powe
r particularly in a situation when the targeted outputs have large var
iations in experimental errors.