Explainable Galaxy Interaction Prediction with Hybrid Attention Mechanisms
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917202504974336 |
|---|---|
| author | Narkedimilli, Sathwik Raghav, Satvik Mishra, Om Kumar, Mohan H, Aswath Babu Jerabkova, Tereza M, Manish Mallellu, Sai Prashanth |
| author_facet | Narkedimilli, Sathwik Raghav, Satvik Mishra, Om Kumar, Mohan H, Aswath Babu Jerabkova, Tereza M, Manish Mallellu, Sai Prashanth |
| contents | Galaxy interaction classification remains challenging due to complex morphological patterns and the limited interpretability of deep learning models. We propose an attentive neural ensemble that combines AG-XCaps, H-SNN, and ResNet-GRU architectures, trained on the Galaxy Zoo DESI dataset and enhanced with LIME to enable explainable predictions. The model achieves Precision = 0.95, Recall = 1.00, F1 = 0.97, and Accuracy = 96%, outperforming a Random Forest baseline by significantly reducing false positives (23 vs. 70). This lightweight (0.45 MB) and scalable framework provides an interpretable and efficient solution for large-scale surveys such as Euclid and LSST, advancing data-driven studies of galaxy evolution. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_08872 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Explainable Galaxy Interaction Prediction with Hybrid Attention Mechanisms Narkedimilli, Sathwik Raghav, Satvik Mishra, Om Kumar, Mohan H, Aswath Babu Jerabkova, Tereza M, Manish Mallellu, Sai Prashanth Instrumentation and Methods for Astrophysics Astrophysics of Galaxies Galaxy interaction classification remains challenging due to complex morphological patterns and the limited interpretability of deep learning models. We propose an attentive neural ensemble that combines AG-XCaps, H-SNN, and ResNet-GRU architectures, trained on the Galaxy Zoo DESI dataset and enhanced with LIME to enable explainable predictions. The model achieves Precision = 0.95, Recall = 1.00, F1 = 0.97, and Accuracy = 96%, outperforming a Random Forest baseline by significantly reducing false positives (23 vs. 70). This lightweight (0.45 MB) and scalable framework provides an interpretable and efficient solution for large-scale surveys such as Euclid and LSST, advancing data-driven studies of galaxy evolution. |
| title | Explainable Galaxy Interaction Prediction with Hybrid Attention Mechanisms |
| topic | Instrumentation and Methods for Astrophysics Astrophysics of Galaxies |
| url | https://arxiv.org/abs/2601.08872 |