Beyond Patches: Mining Interpretable Part-Prototypes for Explainable AI
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917094656835584 |
|---|---|
| author | Alehdaghi, Mahdi Bhattacharya, Rajarshi Shamsolmoali, Pourya Cruz, Rafael M. O. Heritier, Maguelonne Granger, Eric |
| author_facet | Alehdaghi, Mahdi Bhattacharya, Rajarshi Shamsolmoali, Pourya Cruz, Rafael M. O. Heritier, Maguelonne Granger, Eric |
| contents | As AI systems grow more capable, it becomes increasingly important that their decisions remain understandable and aligned with human expectations. A key challenge is the limited interpretability of deep models. Post-hoc methods like GradCAM offer heatmaps but provide limited conceptual insight, while prototype-based approaches offer example-based explanations but often rely on rigid region selection and lack semantic consistency.
To address these limitations, we propose PCMNet, a part-prototypical concept mining network that learns human-comprehensible prototypes from meaningful image regions without additional supervision. By clustering these prototypes into concept groups and extracting concept activation vectors, PCMNet provides structured, concept-level explanations and enhances robustness to occlusion and challenging conditions, which are both critical for building reliable and aligned AI systems.
Experiments across multiple image classification benchmarks show that PCMNet outperforms state-of-the-art methods in interpretability, stability, and robustness. This work contributes to AI alignment by enhancing transparency, controllability, and trustworthiness in AI systems. Our code is available at: https://github.com/alehdaghi/PCMNet. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_12197 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Patches: Mining Interpretable Part-Prototypes for Explainable AI Alehdaghi, Mahdi Bhattacharya, Rajarshi Shamsolmoali, Pourya Cruz, Rafael M. O. Heritier, Maguelonne Granger, Eric Computer Vision and Pattern Recognition As AI systems grow more capable, it becomes increasingly important that their decisions remain understandable and aligned with human expectations. A key challenge is the limited interpretability of deep models. Post-hoc methods like GradCAM offer heatmaps but provide limited conceptual insight, while prototype-based approaches offer example-based explanations but often rely on rigid region selection and lack semantic consistency. To address these limitations, we propose PCMNet, a part-prototypical concept mining network that learns human-comprehensible prototypes from meaningful image regions without additional supervision. By clustering these prototypes into concept groups and extracting concept activation vectors, PCMNet provides structured, concept-level explanations and enhances robustness to occlusion and challenging conditions, which are both critical for building reliable and aligned AI systems. Experiments across multiple image classification benchmarks show that PCMNet outperforms state-of-the-art methods in interpretability, stability, and robustness. This work contributes to AI alignment by enhancing transparency, controllability, and trustworthiness in AI systems. Our code is available at: https://github.com/alehdaghi/PCMNet. |
| title | Beyond Patches: Mining Interpretable Part-Prototypes for Explainable AI |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.12197 |