Pre or Post-Softmax Scores in Gradient-based Attribution Methods, What is Best?

Code

• github.com/mlerma54/adversarial-attacks-on-saliency-maps
  Officialnone★ 0

Abstract

Gradient based attribution methods for neural networks working as classifiers use gradients of network scores. Here we discuss the practical differences between using gradients of pre-softmax scores versus post-softmax scores, and their respective advantages and disadvantages.

Reproductions