EXACT: Towards a platform for empirically benchmarking Machine Learning model explanation methods

TL;DR

This paper introduces EXACT, a benchmarking platform with datasets and metrics for evaluating the quality of machine learning explanation methods, revealing limitations of current approaches and variability across models.

Contribution

The paper presents a standardized benchmarking platform for XAI methods, including datasets with ground truth explanations and novel metrics for assessment.

Findings

Popular XAI methods often perform no better than random baselines.

Explanation quality varies significantly across different model architectures.

Current XAI methods struggle to reliably identify relevant features.

Abstract

The evolving landscape of explainable artificial intelligence (XAI) aims to improve the interpretability of intricate machine learning (ML) models, yet faces challenges in formalisation and empirical validation, being an inherently unsupervised process. In this paper, we bring together various benchmark datasets and novel performance metrics in an initial benchmarking platform, the Explainable AI Comparison Toolkit (EXACT), providing a standardised foundation for evaluating XAI methods. Our datasets incorporate ground truth explanations for class-conditional features, and leveraging novel quantitative metrics, this platform assesses the performance of post-hoc XAI methods in the quality of the explanations they produce. Our recent findings have highlighted the limitations of popular XAI methods, as they often struggle to surpass random baselines, attributing significance to irrelevant features. Moreover, we show the variability in explanations derived from different equally performing model architectures. This initial benchmarking platform therefore aims to allow XAI researchers to test and assure the high quality of their newly developed methods.