Learning rule representations from Boolean data

We discuss a Probably Approximate Correct (PAC) learning paradigm for Boolean formulas, which we call PAC meditation, where the class of formulas to be learnt is not known in advance. We split the building of the hypothesis in various levels of increasing description complexity according to additional inductive biases received at run time. In order to give semantic value to the learnt formulas, the key operational aspect represented is the understandability of formulas, which requires their simplification at any level of description. We deepen this aspect in light of two alternative simplification methods, which we compare through a case study.