Attention-Based Explainability for Structure-Property Relationships
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912546819145728 |
|---|---|
| author | Slautin, Boris N. Pratiush, Utkarsh Liu, Yongtao Funakubo, Hiroshi Shvartsman, Vladimir V. Lupascu, Doru C. Kalinin, Sergei V. |
| author_facet | Slautin, Boris N. Pratiush, Utkarsh Liu, Yongtao Funakubo, Hiroshi Shvartsman, Vladimir V. Lupascu, Doru C. Kalinin, Sergei V. |
| contents | Machine learning methods are emerging as a universal paradigm for constructing correlative structure-property relationships in materials science based on multimodal characterization. However, this necessitates development of methods for physical interpretability of the resulting correlative models. Here, we demonstrate the potential of attention-based neural networks for revealing structure-property relationships and the underlying physical mechanisms, using the ferroelectric properties of PbTiO3 thin films as a case study. Through the analysis of attention scores, we disentangle the influence of distinct domain patterns on the polarization switching process. The attention-based Transformer model is explored both as a direct interpretability tool and as a surrogate for explaining representations learned via unsupervised machine learning, enabling the identification of physically grounded correlations. We compare attention-derived interpretability scores with classical SHapley Additive exPlanations (SHAP) analysis and show that, in contrast to applications in natural language processing, attention mechanisms in materials science exhibit high efficiency in highlighting meaningful structural features. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_15493 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Attention-Based Explainability for Structure-Property Relationships Slautin, Boris N. Pratiush, Utkarsh Liu, Yongtao Funakubo, Hiroshi Shvartsman, Vladimir V. Lupascu, Doru C. Kalinin, Sergei V. Materials Science Machine learning methods are emerging as a universal paradigm for constructing correlative structure-property relationships in materials science based on multimodal characterization. However, this necessitates development of methods for physical interpretability of the resulting correlative models. Here, we demonstrate the potential of attention-based neural networks for revealing structure-property relationships and the underlying physical mechanisms, using the ferroelectric properties of PbTiO3 thin films as a case study. Through the analysis of attention scores, we disentangle the influence of distinct domain patterns on the polarization switching process. The attention-based Transformer model is explored both as a direct interpretability tool and as a surrogate for explaining representations learned via unsupervised machine learning, enabling the identification of physically grounded correlations. We compare attention-derived interpretability scores with classical SHapley Additive exPlanations (SHAP) analysis and show that, in contrast to applications in natural language processing, attention mechanisms in materials science exhibit high efficiency in highlighting meaningful structural features. |
| title | Attention-Based Explainability for Structure-Property Relationships |
| topic | Materials Science |
| url | https://arxiv.org/abs/2508.15493 |