On Interpretability of Artificial Neural Networks: A Survey

Abstract

Deep learning as performed by artificial deep neural networks (DNNs) has achieved great successes recently in many important areas that deal with text, images, videos, graphs, and so on. However, the black-box nature of DNNs has become one of the primary obstacles for their wide adoption in mission-critical applications such as medical diagnosis and therapy. Because of the huge potentials of deep learning, the interpretability of DNNs has recently attracted much research attention. In this article, we propose a simple but comprehensive taxonomy for interpretability, systematically review recent studies on interpretability of neural networks, describe applications of interpretability in medicine, and discuss future research directions, such as in relation to fuzzy logic and brain science.