A Survey on Neural Network Interpretability

TL;DR

This survey reviews neural network interpretability, clarifies its definitions, proposes a novel 3D taxonomy, and discusses evaluation methods, aiming to advance understanding and research in making neural networks more transparent.

Contribution

It introduces a new 3D taxonomy for interpretability approaches and provides a comprehensive overview of existing evaluation methods and future research directions.

Findings

Proposes a novel 3D interpretability taxonomy.

Summarizes existing interpretability evaluation methods.

Suggests future research directions based on taxonomy.

Abstract

Along with the great success of deep neural networks, there is also growing concern about their black-box nature. The interpretability issue affects people's trust on deep learning systems. It is also related to many ethical problems, e.g., algorithmic discrimination. Moreover, interpretability is a desired property for deep networks to become powerful tools in other research fields, e.g., drug discovery and genomics. In this survey, we conduct a comprehensive review of the neural network interpretability research. We first clarify the definition of interpretability as it has been used in many different contexts. Then we elaborate on the importance of interpretability and propose a novel taxonomy organized along three dimensions: type of engagement (passive vs. active interpretation approaches), the type of explanation, and the focus (from local to global interpretability). This taxonomy provides a meaningful 3D view of distribution of papers from the relevant literature as two of the dimensions are not simply categorical but allow ordinal subcategories. Finally, we summarize the existing interpretability evaluation methods and suggest possible research directions inspired by our new taxonomy.