On the Computational Intelligibility of Boolean Classifiers

TL;DR

This paper analyzes the computational difficulty of explaining various Boolean classifiers, showing decision trees are tractable for XAI queries while other models are not, highlighting a significant gap in interpretability.

Contribution

It provides a systematic comparison of the computational intelligibility of different Boolean classifiers across multiple XAI queries, revealing a stark contrast between decision trees and other models.

Findings

Decision trees are tractable for all tested XAI queries.

DNF formulae, decision lists, and neural networks are intractable for these queries.

Large intelligibility gap exists between decision trees and other classifiers.

Abstract

In this paper, we investigate the computational intelligibility of Boolean classifiers, characterized by their ability to answer XAI queries in polynomial time. The classifiers under consideration are decision trees, DNF formulae, decision lists, decision rules, tree ensembles, and Boolean neural nets. Using 9 XAI queries, including both explanation queries and verification queries, we show the existence of large intelligibility gap between the families of classifiers. On the one hand, all the 9 XAI queries are tractable for decision trees. On the other hand, none of them is tractable for DNF formulae, decision lists, random forests, boosted decision trees, Boolean multilayer perceptrons, and binarized neural networks.