Unified Explanations in Machine Learning Models: A Perturbation Approach

TL;DR

This paper introduces a perturbation-based framework to evaluate and compare explanation methods in machine learning, ensuring consistency and reliability of model explanations across different models and datasets.

Contribution

It proposes a systematic perturbation approach to analyze and quantify the alignment of explanation methods like Shap across various models and datasets.

Findings

High similarity in explanations across models and datasets

Quantitative metrics for explanation consistency

A taxonomy for feature importance methods

Abstract

A high-velocity paradigm shift towards Explainable Artificial Intelligence (XAI) has emerged in recent years. Highly complex Machine Learning (ML) models have flourished in many tasks of intelligence, and the questions have started to shift away from traditional metrics of validity towards something deeper: What is this model telling me about my data, and how is it arriving at these conclusions? Inconsistencies between XAI and modeling techniques can have the undesirable effect of casting doubt upon the efficacy of these explainability approaches. To address these problems, we propose a systematic, perturbation-based analysis against a popular, model-agnostic method in XAI, SHapley Additive exPlanations (Shap). We devise algorithms to generate relative feature importance in settings of dynamic inference amongst a suite of popular machine learning and deep learning methods, and metrics that allow us to quantify how well explanations generated under the static case hold. We propose a taxonomy for feature importance methodology, measure alignment, and observe quantifiable similarity amongst explanation models across several datasets.