Towards Desiderata-Driven Design of Visual Counterfactual Explainers

TL;DR

This paper introduces a new 'smooth counterfactual explorer' algorithm for visual counterfactual explanations, aiming to improve explanation quality by considering holistic desiderata like fidelity and understandability, beyond just minimal changes.

Contribution

It proposes a novel mechanism and algorithm for generating visual counterfactual explanations that address holistic desiderata, advancing beyond existing narrow optimization criteria.

Findings

The SCE algorithm effectively improves explanation quality.

Systematic evaluations show better fidelity and understandability.

Demonstrated on synthetic and real data.

Abstract

Visual counterfactual explainers (VCEs) are a straightforward and promising approach to enhancing the transparency of image classifiers. VCEs complement other types of explanations, such as feature attribution, by revealing the specific data transformations to which a machine learning model responds most strongly. In this paper, we argue that existing VCEs focus too narrowly on optimizing sample quality or change minimality; they fail to consider the more holistic desiderata for an explanation, such as fidelity, understandability, and sufficiency. To address this shortcoming, we explore new mechanisms for counterfactual generation and investigate how they can help fulfill these desiderata. We combine these mechanisms into a novel 'smooth counterfactual explorer' (SCE) algorithm and demonstrate its effectiveness through systematic evaluations on synthetic and real data.