Towards improving saliency map interpretability using feature map smoothing

Dipkamal Bhusal, Md Tanvirul Alam, Monish Kumar manikya Veerabhadran, Michael Clifford, Sara Rampazzi, Nidhi Rastogi

Keywords: computer vision; saliency map; explanations; vanilla gradient; integrated gradient; smoothgrad

TL;DR: The paper studies the reason behind stability issues of input-gradient based attribution methods and demonstrates how model robustness with smoothing blocks improve post-hoc explanations of such methods.

Abstract: Input-gradient-based feature attribution methods, such as Vanilla Gradient, Integrated Gradients, and SmoothGrad, are widely used to explain image classifiers by generating saliency maps. However, these methods struggle to provide explanations that are both visually clear and quantitatively robust. Key challenges include ensuring that explanations are sparse, stable, and faithfully reflect the model’s decision-making. Adversarial training, known for enhancing model robustness, have been shown to produce sparser explanations with these methods; however, this sparsity often comes at the cost of stability. In this work, we investigate the trade-off between stability and sparsity in saliency maps and propose the use of a smoothing layer during adversarial training. Through extensive experiments and evaluation, we demonstrate this smoothing technique improves the stability and faithfulness of saliency maps without sacrificing sparsity. Furthermore, a qualitative user study reveals that human evaluators tend to distrust explanations that are overly noisy or excessively sparse—issues commonly associated with explanations in naturally and adversarially trained models, respectively and prefer explanations produced by our proposed approach. Our findings offer a promising direction for generating reliable explanations with robust models, striking a balance between clarity and usability.

Primary Area: interpretability and explainable AI

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 3260