ABSTRACT
This chapter utilizes the regression-based neural network (RBNN) model for solving first- and higher-order ordinary differential equations (ODEs). The trial solution of the differential equation has been obtained by using the RBNN model for single input and single output (SISO) system. The chapter also utilizes the unsupervised error back-propagation learning algorithm to update the network parameters (weights and biases) from the input layer to the hidden layer and from the hidden layer to the output layer and to minimize the error function of the RBNN model. It explains how to modify of parameters without the use of any optimization technique. The chapter solves a variety of initial and boundary value problems and compares the results with arbitrary and regression-based initial weights. It provides examples considering three first-order ODEs to show the reliability of the RBNN model and the accuracy of results of the proposed RBNN model.
