Benchmarking and Survey of Explanation Methods for Black Box Models

TL;DR

This paper provides a comprehensive categorization, comparison, and benchmarking of explanation methods for black-box AI models, highlighting their differences and practical applications.

Contribution

It offers a systematic categorization of explanation methods, presents a visual comparison, and provides a quantitative benchmark of the most recent explainers.

Findings

Categorized explanation methods based on explanation type

Visual comparison of different explanation methods

Quantitative benchmarking of explanation effectiveness

Abstract

The widespread adoption of black-box models in Artificial Intelligence has enhanced the need for explanation methods to reveal how these obscure models reach specific decisions. Retrieving explanations is fundamental to unveil possible biases and to resolve practical or ethical issues. Nowadays, the literature is full of methods with different explanations. We provide a categorization of explanation methods based on the type of explanation returned. We present the most recent and widely used explainers, and we show a visual comparison among explanations and a quantitative benchmarking.