Towards Quantification of Explainability in Explainable Artificial Intelligence Methods
This addresses the problem of standardizing explainability metrics for AI systems, which is crucial for domains like healthcare and finance, but it is incremental as it builds on prior work without introducing a new paradigm.
The paper tackles the challenge of quantifying explainability in Explainable AI (XAI) by proposing a model-agnostic approach to measure the extent of explainability, with experimental results suggesting a reasonable method for this quantification.
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