EDDA: Explanation-driven Data Augmentation to Improve Explanation Faithfulness

TL;DR

This paper introduces EDDA, a data augmentation method that enhances the faithfulness of explanations for image classifiers by using occlusion-based augmentation driven by explanation methods, improving model debugging.

Contribution

The paper proposes a novel explanation-driven data augmentation technique that improves explanation faithfulness without requiring ground truth explanations.

Findings

EDDA significantly increases explanation faithfulness across datasets.

Augmentation improves model debugging and deployment.

Method is effective with various models and datasets.

Abstract

Recent years have seen the introduction of a range of methods for post-hoc explainability of image classifier predictions. However, these post-hoc explanations may not always be faithful to classifier predictions, which poses a significant challenge when attempting to debug models based on such explanations. To this end, we seek a methodology that can improve the faithfulness of an explanation method with respect to model predictions which does not require ground truth explanations. We achieve this through a novel explanation-driven data augmentation (EDDA) technique that augments the training data with occlusions inferred from model explanations; this is based on the simple motivating principle that \emph{if} the explainer is faithful to the model \emph{then} occluding salient regions for the model prediction should decrease the model confidence in the prediction, while occluding non-salient regions should not change the prediction. To verify that the proposed augmentation method has the potential to improve faithfulness, we evaluate EDDA using a variety of datasets and classification models. We demonstrate empirically that our approach leads to a significant increase of faithfulness, which can facilitate better debugging and successful deployment of image classification models in real-world applications.