Inverse Classification for Comparison-based Interpretability in Machine Learning

TL;DR

This paper introduces an instance-based method for post-hoc interpretability that explains classifier predictions by finding minimal changes needed to flip the classification, without requiring access to the classifier or data.

Contribution

It proposes a novel approach using Growing Spheres to generate close neighbors with different classifications, enhancing understanding of black-box models.

Findings

Effective in explaining predictions without classifier access

Applicable to different datasets demonstrating versatility

Provides insights into decision boundaries

Abstract

In the context of post-hoc interpretability, this paper addresses the task of explaining the prediction of a classifier, considering the case where no information is available, neither on the classifier itself, nor on the processed data (neither the training nor the test data). It proposes an instance-based approach whose principle consists in determining the minimal changes needed to alter a prediction: given a data point whose classification must be explained, the proposed method consists in identifying a close neighbour classified differently, where the closeness definition integrates a sparsity constraint. This principle is implemented using observation generation in the Growing Spheres algorithm. Experimental results on two datasets illustrate the relevance of the proposed approach that can be used to gain knowledge about the classifier.