SAT-Based Rigorous Explanations for Decision Lists

TL;DR

This paper demonstrates that computing explanations for decision lists is computationally hard, and proposes SAT-based methods for efficient, rigorous explanations, showing practical feasibility in real-world scenarios.

Contribution

It introduces propositional encodings for abductive and contrastive explanations of decision lists and evaluates their practical efficiency using SAT solvers.

Findings

SAT-based explanations are computationally feasible for practical decision lists.

Complete enumeration of explanations is often achievable in real settings.

Propositional encodings enable rigorous explanations for decision lists.

Abstract

Decision lists (DLs) find a wide range of uses for classification problems in Machine Learning (ML), being implemented in a number of ML frameworks. DLs are often perceived as interpretable. However, building on recent results for decision trees (DTs), we argue that interpretability is an elusive goal for some DLs. As a result, for some uses of DLs, it will be important to compute (rigorous) explanations. Unfortunately, and in clear contrast with the case of DTs, this paper shows that computing explanations for DLs is computationally hard. Motivated by this result, the paper proposes propositional encodings for computing abductive explanations (AXps) and contrastive explanations (CXps) of DLs. Furthermore, the paper investigates the practical efficiency of a MARCO-like approach for enumerating explanations. The experimental results demonstrate that, for DLs used in practical settings, the use of SAT oracles offers a very efficient solution, and that complete enumeration of explanations is most often feasible.