Explainable Global Fairness Verification of Tree-Based Classifiers

TL;DR

This paper introduces a formal, explainable method for globally verifying fairness in tree-based classifiers by synthesizing human-understandable logical conditions that guarantee fairness across all inputs.

Contribution

It provides a sound and complete approach to generate global fairness guarantees as propositional logic formulas for tree-based classifiers.

Findings

The method is precise and explainable to human experts.

It is efficient enough for practical use.

Experimental results validate the approach on public datasets.

Abstract

We present a new approach to the global fairness verification of tree-based classifiers. Given a tree-based classifier and a set of sensitive features potentially leading to discrimination, our analysis synthesizes sufficient conditions for fairness, expressed as a set of traditional propositional logic formulas, which are readily understandable by human experts. The verified fairness guarantees are global, in that the formulas predicate over all the possible inputs of the classifier, rather than just a few specific test instances. Our analysis is formally proved both sound and complete. Experimental results on public datasets show that the analysis is precise, explainable to human experts and efficient enough for practical adoption.