Towards Quantification of Explainability in Explainable Artificial Intelligence Methods

TL;DR

This paper explores the challenge of quantifying explainability in AI, analyzing interdisciplinary perspectives and proposing a model-agnostic approach to measure how understandable AI decisions are.

Contribution

It introduces a formal method to quantify explainability in AI, bridging gaps across disciplines and providing a practical, model-agnostic measurement approach.

Findings

Proposed a formal approach to quantify explainability

Analyzed interdisciplinary perspectives on explainability

Suggested a model-agnostic quantification method

Abstract

Artificial Intelligence (AI) has become an integral part of domains such as security, finance, healthcare, medicine, and criminal justice. Explaining the decisions of AI systems in human terms is a key challenge--due to the high complexity of the model, as well as the potential implications on human interests, rights, and lives . While Explainable AI is an emerging field of research, there is no consensus on the definition, quantification, and formalization of explainability. In fact, the quantification of explainability is an open challenge. In our previous work, we incorporated domain knowledge for better explainability, however, we were unable to quantify the extent of explainability. In this work, we (1) briefly analyze the definitions of explainability from the perspective of different disciplines (e.g., psychology, social science), properties of explanation, explanation methods, and human-friendly explanations; and (2) propose and formulate an approach to quantify the extent of explainability. Our experimental result suggests a reasonable and model-agnostic way to quantify explainability