Beyond the Veil of Similarity: Quantifying Semantic Continuity in Explainable AI

TL;DR

This paper proposes a new metric to quantify semantic continuity in explainable AI, assessing how explanations change with incremental input variations to improve interpretability and trustworthiness.

Contribution

It introduces a novel quantitative measure for semantic continuity in XAI, enabling evaluation of explanation consistency and model understanding.

Findings

Semantic continuity correlates with model generalization.

Different XAI methods vary in capturing semantic consistency.

The metric helps identify more reliable explanation techniques.

Abstract

We introduce a novel metric for measuring semantic continuity in Explainable AI methods and machine learning models. We posit that for models to be truly interpretable and trustworthy, similar inputs should yield similar explanations, reflecting a consistent semantic understanding. By leveraging XAI techniques, we assess semantic continuity in the task of image recognition. We conduct experiments to observe how incremental changes in input affect the explanations provided by different XAI methods. Through this approach, we aim to evaluate the models' capability to generalize and abstract semantic concepts accurately and to evaluate different XAI methods in correctly capturing the model behaviour. This paper contributes to the broader discourse on AI interpretability by proposing a quantitative measure for semantic continuity for XAI methods, offering insights into the models' and explainers' internal reasoning processes, and promoting more reliable and transparent AI systems.