We investigate the network complexity of multilayered perceptrons for solving exactly a given problem. We limit our study to the class of combinatorial optimization problems. It is shown how these problems can be reformulated as binary classification problems and how they can be solved by multilayered perceptrons.
Zwietering, P. J., Aarts, E. H. L., & Wessels, J. (1991). The design and complexity of exact multilayered perceptrons. International Journal of Neural Systems, 2(3), 185-199. https://doi.org/10.1142/S0129065791000170