A Survey of Explainable AI in Deep Visual Modeling: Methods and Metrics

TL;DR

This survey reviews methods and metrics for interpreting deep visual models in explainable AI, highlighting current trends, challenges, and future research directions in a rapidly evolving field.

Contribution

It provides a comprehensive taxonomic organization of existing interpretability techniques and evaluation metrics for deep visual models.

Findings

Taxonomy of interpretability methods

Compilation of evaluation metrics

Discussion on current challenges and future directions

Abstract

Deep visual models have widespread applications in high-stake domains. Hence, their black-box nature is currently attracting a large interest of the research community. We present the first survey in Explainable AI that focuses on the methods and metrics for interpreting deep visual models. Covering the landmark contributions along the state-of-the-art, we not only provide a taxonomic organization of the existing techniques, but also excavate a range of evaluation metrics and collate them as measures of different properties of model explanations. Along the insightful discussion on the current trends, we also discuss the challenges and future avenues for this research direction.