Easily Reconstructable Functions

This chapter shows that simplification of SOPs gives the generalization ability. We show this in three steps. First, various classes of totally defined logic functions (SOPs) were generated. Second, minterms of the function are randomly selected to generate partially defined functions. And, third, from the partially defined functions, original functions are reconstructed by SOP minimization. We consider the Achilles heel functions, majority functions, monotone increasing cascade functions, functions generated from random SOPs, and monotone increasing random SOPs. As for machine learning methods, Naive Bayes, multi-level perceptron, support vector machine, JRIP, J48, and random forest are considered in addition to SOP minimization.