BTSbot:一种用于自动化并加速兹威基瞬变设施亮瞬变源识别的多输入卷积神经网络
天体物理仪器与方法
2025-06-05 v2
摘要
亮瞬变源巡天(BTS)依赖人工目视检查(“扫描”)来筛选源,以完成其对兹威基瞬变设施(ZTF)发现的所有亮河外瞬变源进行光谱分类的任务。我们提出 BTSbot,一种多输入卷积神经网络,它利用图像数据和14个提取特征为单个ZTF探测提供亮瞬变源评分。BTSbot通过自动识别新的亮瞬变源候选体(m < 18.5 mag)并请求后续观测,消除了扫描的需要。BTSbot在完备性(99% 对比 95%)和识别速度(平均快7.4小时)上均优于BTS扫描人员。完整的BTSbot论文见 Rehemtulla 等人 2024, ApJ, 972, 7R。
引用
@article{arxiv.2307.07618,
title = {$\texttt{BTSbot}$: A Multi-input Convolutional Neural Network to Automate and Expedite Bright Transient Identification for the Zwicky Transient Facility},
author = {Nabeel Rehemtulla and Adam A. Miller and Michael W. Coughlin and Theophile Jegou du Laz},
journal= {arXiv preprint arXiv:2307.07618},
year = {2025}
}
备注
Accepted at the ICML 2023 Workshop on ML for Astrophysics; see Rehemtulla et al. 2024, ApJ, 972, 7R for the full BTSbot publication