In a previous paper (Russell et. al., 2002), the authors applied the multi-layer perceptron (MLP) neural network to the solution of a straightforward AVO classification problem. In this paper, we will summarize the earlier work, and then show how the problem can be solved in an alternate fashion by using the radial basis function neural network (RBFN). By using an AVO model as our input, we are able to gain greater insight into the inner workings of neural networks.
View full article as PDF (0.37 Mb)