Explaining text classifiers through progressive neighborhood approximation with realistic samples

TL;DR

This paper introduces a progressive neighborhood approximation method for local text classifier explanations, utilizing counterfactuals and realistic samples to improve interpretability and fidelity.

Contribution

It proposes a novel two-stage interpolation approach and a probability-based alternative for generating realistic neighborhoods in local explanations of text classifiers.

Findings

Both methods produce more realistic and interpretable explanations.

Experimental results show improved explanation quality over existing approaches.

The approaches are effective across various datasets and models.

Abstract

The importance of neighborhood construction in local explanation methods has been already highlighted in the literature. And several attempts have been made to improve neighborhood quality for high-dimensional data, for example, texts, by adopting generative models. Although the generators produce more realistic samples, the intuitive sampling approaches in the existing solutions leave the latent space underexplored. To overcome this problem, our work, focusing on local model-agnostic explanations for text classifiers, proposes a progressive approximation approach that refines the neighborhood of a to-be-explained decision with a careful two-stage interpolation using counterfactuals as landmarks. We explicitly specify the two properties that should be satisfied by generative models, the reconstruction ability and the locality-preserving property, to guide the selection of generators for local explanation methods. Moreover, noticing the opacity of generative models during the study, we propose another method that implements progressive neighborhood approximation with probability-based editions as an alternative to the generator-based solution. The explanation results from both methods consist of word-level and instance-level explanations benefiting from the realistic neighborhood. Through exhaustive experiments, we qualitatively and quantitatively demonstrate the effectiveness of the two proposed methods.