Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability
This work addresses the problem of enhancing reliability and interpretability for users of deep neural networks, though it appears incremental as it builds on existing attribution methods with frequency-aware attacks.
The paper tackles the challenge of improving explainability for deep neural networks under noise and perturbations by proposing a new attribution method called Frequency-Aware Model Parameter Explorer (FAMPE), which achieves an average gain of 13.02% in Insertion Score compared to the state-of-the-art method AttEXplore.
Ensuring the reliability of deep neural networks (DNNs) in the presence of real world noise and intentional perturbations remains a significant challenge. To address this, attribution methods have been proposed, though their efficacy remains suboptimal and necessitates further refinement. In this paper, we propose a novel category of transferable adversarial attacks, called transferable frequency-aware attacks, enabling frequency-aware exploration via both high-and low-frequency components. Based on this type of attacks, we also propose a novel attribution method, named Frequency-Aware Model Parameter Explorer (FAMPE), which improves the explainability for DNNs. Relative to the current state-of-the-art method AttEXplore, our FAMPE attains an average gain of 13.02% in Insertion Score, thereby outperforming existing approaches. Through detailed ablation studies, we also investigate the role of both high- and low-frequency components in explainability.