SAFE: Saliency-Aware Counterfactual Explanations for DNN-based Automated Driving Systems

TL;DR

This paper introduces a saliency-aware counterfactual explanation method for deep neural network-based automated driving systems, aiming to generate more accurate and boundary-close counterfactuals by leveraging saliency maps.

Contribution

The paper proposes a novel approach that uses saliency maps to improve the quality of counterfactual explanations in DNNs, addressing limitations of existing generative models.

Findings

Produces more informative counterfactuals near decision boundaries

Addresses the discrepancy between feature selection and decision boundary proximity

Enhances interpretability of DNN decisions in automated driving systems

Abstract

A CF explainer identifies the minimum modifications in the input that would alter the model's output to its complement. In other words, a CF explainer computes the minimum modifications required to cross the model's decision boundary. Current deep generative CF models often work with user-selected features rather than focusing on the discriminative features of the black-box model. Consequently, such CF examples may not necessarily lie near the decision boundary, thereby contradicting the definition of CFs. To address this issue, we propose in this paper a novel approach that leverages saliency maps to generate more informative CF explanations. Source codes are available at: https://github.com/Amir-Samadi//Saliency_Aware_CF.