DisCERN:Discovering Counterfactual Explanations using Relevance Features from Neighbourhoods

TL;DR

DisCERN is a case-based counterfactual explanation method that uses relevance features from neighborhoods to identify minimal actionable feature changes, outperforming optimization-based approaches in efficiency.

Contribution

This paper introduces DisCERN, a novel counterfactual explainer leveraging relevance features and nearest unlike neighbors to efficiently find minimal actionable changes.

Findings

DisCERN effectively minimizes feature changes for counterfactuals.

DisCERN outperforms DiCE in a comparative study.

Relevance features from LIME and SHAP enhance DisCERN's explanations.

Abstract

Counterfactual explanations focus on "actionable knowledge" to help end-users understand how a machine learning outcome could be changed to a more desirable outcome. For this purpose a counterfactual explainer needs to discover input dependencies that relate to outcome changes. Identifying the minimum subset of feature changes needed to action an output change in the decision is an interesting challenge for counterfactual explainers. The DisCERN algorithm introduced in this paper is a case-based counter-factual explainer. Here counterfactuals are formed by replacing feature values from a nearest unlike neighbour (NUN) until an actionable change is observed. We show how widely adopted feature relevance-based explainers (i.e. LIME, SHAP), can inform DisCERN to identify the minimum subset of "actionable features". We demonstrate our DisCERN algorithm on five datasets in a comparative study with the widely used optimisation-based counterfactual approach DiCE. Our results demonstrate that DisCERN is an effective strategy to minimise actionable changes necessary to create good counterfactual explanations.