Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.03245 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910201074941952 |
|---|---|
| author | Yavari, Ali Mohamadi, Alireza Beydaghi, Elham Seeböck, Philipp Leitgeb, Rainer A. |
| author_facet | Yavari, Ali Mohamadi, Alireza Beydaghi, Elham Seeböck, Philipp Leitgeb, Rainer A. |
| contents | State-of-the-art attribution methods rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum, discarding fine-grained high-frequency information that is demonstrably important for accurate feature attribution in deep neural networks. By generating adversarial samples that selectively perturb high- and low-frequency components, we can probe which spectral features a model relies on most -- directly translating frequency-domain exploration into attribution signals. Building on this insight, we propose FAMPE (Frequency-Aware Model Parameter Explorer), a novel attribution method that introduces an FFT-based α-weighted perturbation scheme -- separately modulating high- and low-frequency components via an energy-driven spectral cutoff -- and, crucially, integrates this frequency-aware exploration directly into model parameter exploration for attribution, a connection that has not been established in prior work. Unlike prior frequency-aware adversarial approaches that target transferability or imperceptibility, FAMPE's specific formulation is designed and validated exclusively for explainability, translating spectral structure into fine-grained attribution maps without requiring any manual baseline selection. Evaluated on ImageNet across four architectures spanning CNNs and Vision Transformers, at fixed α= 0.1 FAMPE outperforms AttEXplore by 4.25% on Inception-v3 and 12.04% on MaxViT-T, with per-sample oracle selection further revealing that low-frequency-dominated images systematically benefit from high-frequency perturbations -- underscoring the potential of adaptive spectral exploration. Our ablation studies confirm that high-frequency perturbations are disproportionately responsible for attribution precision, while excessive low-frequency noise degrades global structural coherence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_03245 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability Yavari, Ali Mohamadi, Alireza Beydaghi, Elham Seeböck, Philipp Leitgeb, Rainer A. Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition State-of-the-art attribution methods rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum, discarding fine-grained high-frequency information that is demonstrably important for accurate feature attribution in deep neural networks. By generating adversarial samples that selectively perturb high- and low-frequency components, we can probe which spectral features a model relies on most -- directly translating frequency-domain exploration into attribution signals. Building on this insight, we propose FAMPE (Frequency-Aware Model Parameter Explorer), a novel attribution method that introduces an FFT-based α-weighted perturbation scheme -- separately modulating high- and low-frequency components via an energy-driven spectral cutoff -- and, crucially, integrates this frequency-aware exploration directly into model parameter exploration for attribution, a connection that has not been established in prior work. Unlike prior frequency-aware adversarial approaches that target transferability or imperceptibility, FAMPE's specific formulation is designed and validated exclusively for explainability, translating spectral structure into fine-grained attribution maps without requiring any manual baseline selection. Evaluated on ImageNet across four architectures spanning CNNs and Vision Transformers, at fixed α= 0.1 FAMPE outperforms AttEXplore by 4.25% on Inception-v3 and 12.04% on MaxViT-T, with per-sample oracle selection further revealing that low-frequency-dominated images systematically benefit from high-frequency perturbations -- underscoring the potential of adaptive spectral exploration. Our ablation studies confirm that high-frequency perturbations are disproportionately responsible for attribution precision, while excessive low-frequency noise degrades global structural coherence. |
| title | Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.03245 |