A NEURAL-NETWORK APPROACH TO GEOSTATISTICAL SIMULATION

Neural networks offer a non-algorithmic approach to geostatistical sim ulation with the possibility of automatic recognition of correlation s tructure. The paper gives a brief overview of neural networks and desc ribes a feedforward, back-propagation network for geostatistical simul ation. The operation of the network is illustrated with two simple one -dimensional examples which can be followed through with hand calculat ions to give an insight into the operation of the network. The converg ence of the network is described in terms of the variogram calculated from the values at each of the output nodes at each iteration.