A Benchmark for Interpretability Methods in Deep Neural Networks

Code

• github.com/google-research/google-research
  Officialtf★ 37,519
• github.com/edpclau/ROAR-for-Scikit
  none★ 0
• github.com/LLNL/fastcam
  pytorch★ 0

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 importance. Only certain ensemble based approaches---VarGrad and SmoothGrad-Squared---outperform such a random assignment of importance. The manner of ensembling remains critical, we show that some approaches do no better then the underlying method but carry a far higher computational burden.

Tasks

Reproductions