A Benchmark for Interpretability Methods in Deep Neural Networks

Sara Hooker
Sara Hooker
Pieter-Jan Kindermans
Pieter-Jan Kindermans

ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), pp. 9734-9745, 2019.

Cited by: 42|Bibtex|Views132
EI
Other Links: academic.microsoft.com|dblp.uni-trier.de

Abstract:

We propose an empirical measure of the approximate accuracy of feature importance estimates in deep neural networks. Our results across several large-scale image classification datasets show that many popular interpretability methods produce estimates of feature importance that are not better than a random designation of feature importanc...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments