Trustworthy AI-based crack-tip segmentation using domain-guided explanations
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915769182322688 |
|---|---|
| author | Talies, Jesco Breitbarth, Eric Melching, David |
| author_facet | Talies, Jesco Breitbarth, Eric Melching, David |
| contents | Ensuring the trustworthiness and robustness of deep learning models remains a fundamental challenge, particularly in high-stakes scientific applications. In this study, we present a framework called attention-guided training that combines explainable artificial intelligence techniques with quantitative evaluation and domain-specific priors to guide model attention. We demonstrate that domain-specific feedback on model explanations during training can enhance the model's generalization capabilities. We validate our approach on the task of semantic crack tip segmentation in digital image correlation data, which is a key application in the fracture mechanical characterization of materials. By aligning model attention with physically meaningful stress fields, such as those described by Williams' analytical solution, attention-guided training ensures that the model focuses on physically relevant regions. This finally leads to improved generalization and more faithful explanations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_20658 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Trustworthy AI-based crack-tip segmentation using domain-guided explanations Talies, Jesco Breitbarth, Eric Melching, David Materials Science Machine Learning Ensuring the trustworthiness and robustness of deep learning models remains a fundamental challenge, particularly in high-stakes scientific applications. In this study, we present a framework called attention-guided training that combines explainable artificial intelligence techniques with quantitative evaluation and domain-specific priors to guide model attention. We demonstrate that domain-specific feedback on model explanations during training can enhance the model's generalization capabilities. We validate our approach on the task of semantic crack tip segmentation in digital image correlation data, which is a key application in the fracture mechanical characterization of materials. By aligning model attention with physically meaningful stress fields, such as those described by Williams' analytical solution, attention-guided training ensures that the model focuses on physically relevant regions. This finally leads to improved generalization and more faithful explanations. |
| title | Trustworthy AI-based crack-tip segmentation using domain-guided explanations |
| topic | Materials Science Machine Learning |
| url | https://arxiv.org/abs/2507.20658 |