Toward Faithful Explanations in Acoustic Anomaly Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917209608028160 |
|---|---|
| author | Elrashid, Maab Deschênes, Anthony Subakan, Cem Ravanelli, Mirco Georges, Rémi Morin, Michael |
| author_facet | Elrashid, Maab Deschênes, Anthony Subakan, Cem Ravanelli, Mirco Georges, Rémi Morin, Michael |
| contents | Interpretability is essential for user trust in real-world anomaly detection applications. However, deep learning models, despite their strong performance, often lack transparency. In this work, we study the interpretability of autoencoder-based models for audio anomaly detection, by comparing a standard autoencoder (AE) with a mask autoencoder (MAE) in terms of detection performance and interpretability. We applied several attribution methods, including error maps, saliency maps, SmoothGrad, Integrated Gradients, GradSHAP, and Grad-CAM. Although MAE shows a slightly lower detection, it consistently provides more faithful and temporally precise explanations, suggesting a better alignment with true anomalies. To assess the relevance of the regions highlighted by the explanation method, we propose a perturbation-based faithfulness metric that replaces them with their reconstructions to simulate normal input. Our findings, based on experiments in a real industrial scenario, highlight the importance of incorporating interpretability into anomaly detection pipelines and show that masked training improves explanation quality without compromising performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12660 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Toward Faithful Explanations in Acoustic Anomaly Detection Elrashid, Maab Deschênes, Anthony Subakan, Cem Ravanelli, Mirco Georges, Rémi Morin, Michael Sound Machine Learning Audio and Speech Processing Interpretability is essential for user trust in real-world anomaly detection applications. However, deep learning models, despite their strong performance, often lack transparency. In this work, we study the interpretability of autoencoder-based models for audio anomaly detection, by comparing a standard autoencoder (AE) with a mask autoencoder (MAE) in terms of detection performance and interpretability. We applied several attribution methods, including error maps, saliency maps, SmoothGrad, Integrated Gradients, GradSHAP, and Grad-CAM. Although MAE shows a slightly lower detection, it consistently provides more faithful and temporally precise explanations, suggesting a better alignment with true anomalies. To assess the relevance of the regions highlighted by the explanation method, we propose a perturbation-based faithfulness metric that replaces them with their reconstructions to simulate normal input. Our findings, based on experiments in a real industrial scenario, highlight the importance of incorporating interpretability into anomaly detection pipelines and show that masked training improves explanation quality without compromising performance. |
| title | Toward Faithful Explanations in Acoustic Anomaly Detection |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2601.12660 |