The (Un)reliability of saliency methods

11/02/2017
by   Pieter-Jan Kindermans, et al.
0

Saliency methods aim to explain the predictions of deep neural networks. These methods lack reliability when the explanation is sensitive to factors that do not contribute to the model prediction. We use a simple and common pre-processing step ---adding a constant shift to the input data--- to show that a transformation with no effect on the model can cause numerous methods to incorrectly attribute. In order to guarantee reliability, we posit that methods should fulfill input invariance, the requirement that a saliency method mirror the sensitivity of the model with respect to transformations of the input. We show, through several examples, that saliency methods that do not satisfy input invariance result in misleading attribution.

READ FULL TEXT
research
09/20/2023

COSE: A Consistency-Sensitivity Metric for Saliency on Image Classification

We present a set of metrics that utilize vision priors to effectively as...
research
10/12/2020

The elephant in the interpretability room: Why use attention as explanation when we have saliency methods?

There is a recent surge of interest in using attention as explanation of...
research
08/07/2023

On genuine invariance learning without weight-tying

In this paper, we investigate properties and limitations of invariance l...
research
05/13/2021

Sanity Simulations for Saliency Methods

Saliency methods are a popular class of feature attribution tools that a...
research
06/08/2018

Noise-adding Methods of Saliency Map as Series of Higher Order Partial Derivative

SmoothGrad and VarGrad are techniques that enhance the empirical quality...
research
06/07/2022

Beyond Faithfulness: A Framework to Characterize and Compare Saliency Methods

Saliency methods calculate how important each input feature is to a mach...
research
05/27/2019

A Simple Saliency Method That Passes the Sanity Checks

There is great interest in *saliency methods* (also called *attribution ...

Please sign up or login with your details

Forgot password? Click here to reset