dorsal/arxiv
View SchemaExplainable Galaxy Interaction Prediction with Hybrid Attention Mechanisms
| Authors | Sathwik Narkedimilli, Satvik Raghav, Om Mishra, Mohan Kumar, Aswath Babu H, Tereza Jerabkova, Manish M, Sai Prashanth Mallellu |
|---|---|
| Categories | |
| ArXiv ID | 2601.08872vv1 |
| URL | https://arxiv.org/abs/2601.08872 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
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.
{
"annotation_id": "f0c4a476-3113-45a3-84d8-00ca600e0d97",
"date_created": "2026-02-17T05:53:20.166000Z",
"date_modified": "2026-02-17T05:53:20.166000Z",
"file_hash": "10d02c554cef9d2de15cb2ee929b50aa7bad1ba1ac3a5ceac2f1a064f9baddff",
"private": false,
"record": {
"abstract": "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.",
"arxiv_id": "2601.08872",
"authors": [
"Sathwik Narkedimilli",
"Satvik Raghav",
"Om Mishra",
"Mohan Kumar",
"Aswath Babu H",
"Tereza Jerabkova",
"Manish M",
"Sai Prashanth Mallellu"
],
"categories": [
"astro-ph.IM",
"astro-ph.GA"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Explainable Galaxy Interaction Prediction with Hybrid Attention Mechanisms",
"url": "https://arxiv.org/abs/2601.08872",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "1cfb5d06-0e79-44d2-91af-186c1a7b2e65",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}