Debiased-CAM to mitigate image perturbations with faithful visual explanations of machine learning

TL;DR

This paper introduces Debiased-CAM, a method that improves the faithfulness of visual explanations in AI models by reducing bias effects, leading to more accurate and trustworthy saliency maps across various image perturbations.

Contribution

Debiased-CAM is a novel multi-task training approach that enhances explanation faithfulness and prediction accuracy despite systematic image biases.

Findings

Enhanced prediction accuracy across biased images

Generated highly faithful explanations similar to unbiased images

Improved user trust and task performance with debiased explanations

Abstract

Model explanations such as saliency maps can improve user trust in AI by highlighting important features for a prediction. However, these become distorted and misleading when explaining predictions of images that are subject to systematic error (bias) by perturbations and corruptions. Furthermore, the distortions persist despite model fine-tuning on images biased by different factors (blur, color temperature, day/night). We present Debiased-CAM to recover explanation faithfulness across various bias types and levels by training a multi-input, multi-task model with auxiliary tasks for explanation and bias level predictions. In simulation studies, the approach not only enhanced prediction accuracy, but also generated highly faithful explanations about these predictions as if the images were unbiased. In user studies, debiased explanations improved user task performance, perceived truthfulness and perceived helpfulness. Debiased training can provide a versatile platform for robust performance and explanation faithfulness for a wide range of applications with data biases.