ABSTRACT
Zhu et al., 1998). In this study, the feasibility of using artificial neural network in developing accurate and parsimonious constitutive models for ballast behavior is investigated.
2 BRIEF OVERVIEW OF ARTIFICIAL NEURAL NETWORKS
The type of artificial neural networks (ANNs) used in this study are multilayer perceptrons (MLPs) that are trained with the back-propagation algorithm (Rumelhart et al., 1986). A comprehensive description of back-propagation MLPs is beyond the scope of this paper but can be found in Fausett (1994).
