REPEAT: Improving Uncertainty Estimation in Representation Learning Explainability

TL;DR

REPEAT introduces a novel approach to uncertainty estimation in representation learning explainable AI by modeling pixel importance as Bernoulli variables, providing more intuitive and reliable certainty estimates.

Contribution

It proposes a new method called REPEAT that effectively estimates pixel importance certainty by leveraging stochasticity in existing R-XAI techniques.

Findings

REPEAT produces more intuitive certainty estimates.

It improves out-of-distribution detection.

The method yields more concise importance explanations.

Abstract

Incorporating uncertainty is crucial to provide trustworthy explanations of deep learning models. Recent works have demonstrated how uncertainty modeling can be particularly important in the unsupervised field of representation learning explainable artificial intelligence (R-XAI). Current R-XAI methods provide uncertainty by measuring variability in the importance score. However, they fail to provide meaningful estimates of whether a pixel is certainly important or not. In this work, we propose a new R-XAI method called REPEAT that addresses the key question of whether or not a pixel is \textit{certainly} important. REPEAT leverages the stochasticity of current R-XAI methods to produce multiple estimates of importance, thus considering each pixel in an image as a Bernoulli random variable that is either important or unimportant. From these Bernoulli random variables we can directly estimate the importance of a pixel and its associated certainty, thus enabling users to determine certainty in pixel importance. Our extensive evaluation shows that REPEAT gives certainty estimates that are more intuitive, better at detecting out-of-distribution data, and more concise.