Computer Graphics Laboratory ETH Zurich

ETH

Gradient-Based Attribution Methods

M. Ancona, E. Ceolini, A. C. Öztireli, M. Gross

Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer, 2019

Abstract

The problem of explaining complex machine learning models, including Deep Neural Networks, has gained increasing attention over the last few years. While several methods have been proposed to explain network predictions, the definition itself of explanation is still debated. Moreover, only a few attempts to compare explanation methods from a theoretical perspective has been done. In this chapter, we discuss the theoretical properties of several attribution methods and show how they share the same idea of using the gradient information as a descriptive factor for the functioning of a model. Finally, we discuss the strengths and limitations of these methods and compare them with available alternatives.

Downloads

Download Paper
[PDF]