Explaining Naive Bayes and Other Linear Classifiers with Polynomial Time and Delay

TL;DR

This paper introduces efficient algorithms for computing PI-explanations of Naive Bayes and linear classifiers, achieving polynomial delay and log-linear time complexity, significantly improving over previous exponential-time methods.

Contribution

The authors present the first polynomial-time algorithms for computing and enumerating PI-explanations of Naive Bayes and linear classifiers, with experimental validation.

Findings

Algorithms achieve log-linear time for single PI-explanation

Enumeration of explanations with polynomial delay

Experimental results show significant performance improvements

Abstract

Recent work proposed the computation of so-called PI-explanations of Naive Bayes Classifiers (NBCs). PI-explanations are subset-minimal sets of feature-value pairs that are sufficient for the prediction, and have been computed with state-of-the-art exact algorithms that are worst-case exponential in time and space. In contrast, we show that the computation of one PI-explanation for an NBC can be achieved in log-linear time, and that the same result also applies to the more general class of linear classifiers. Furthermore, we show that the enumeration of PI-explanations can be obtained with polynomial delay. Experimental results demonstrate the performance gains of the new algorithms when compared with earlier work. The experimental results also investigate ways to measure the quality of heuristic explanations