Online Learning of k-CNF Boolean Functions

Abstract

This paper revisits the problem of learning a k-CNF Boolean function from examples in the context of online learning under the logarithmic loss. In doing so, we give a Bayesian interpretation to one of Valiant's celebrated PAC learning algorithms, which we then build upon to derive two efficient, online, probabilistic, supervised learning algorithms for predicting the output of an unknown k-CNF Boolean function. We analyze the loss of our methods, and show that the cumulative log-loss can be upper bounded, ignoring logarithmic factors, by a polynomial function of the size of each example.

Using Background Knowledge and Random Sampling in Genetic Programming: A Case Study in Learning Boolean Parity Functions

“…Earlier works on learning Boolean functions include learning of conjunctive normal form from examples (Hirschberg et al, 1994). Learning Boolean functions from examples has continued to be an important problem in machine learning (Veness and Hutter, 2014).…”