arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.03535astro-ph.IMcs.AI
收藏

螳螂虾:一个多巡天计算机视觉光度红移模型的初步报告

Preliminary Report on Mantis Shrimp: a Multi-Survey Computer Vision Photometric Redshift Model

Pacific Northwest National Laboratory(西北太平洋国家实验室);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);National Center for Supercomputing Applications(国家超级计算应用中心);Illinois Center for Advanced Studies of the Universe(伊利诺伊大学宇宙高级研究中心)

Andrew Engel, Gautham Narayan, Nell Byler

AI总结:

本报告介绍螳螂虾模型,一种融合紫外、光学和红外多巡天图像的计算机视觉光度红移估计方法,利用可解释性诊断分析CNN行为,并结合星系物理知识解释模型决策。

摘要:

大型、公开、多模态天文数据集的可用性为开展跨越人工智能科学与天文学科学界限的新颖研究提供了机会。光度红移估计是天文学中一个成熟的分支领域。先前的研究表明,计算机视觉模型通常优于基于星表的模型,但这些模型在整合来自多个仪器或传感器的图像时会面临额外的复杂性。在本报告中,我们详细介绍了构建螳螂虾(Mantis Shrimp)的进展,这是一个用于光度红移估计的多巡天计算机视觉模型,融合了紫外(GALEX)、光学(PanSTARRS)和红外(UnWISE)图像。我们使用深度学习可解释性诊断来衡量模型如何利用不同输入的信息。我们从可解释性指标出发,对卷积神经网络(CNN)的行为进行推理,并特别将结果置于基于物理的星系属性知识框架中加以阐述。

英文摘要:

The availability of large, public, multi-modal astronomical datasets presents an opportunity to execute novel research that straddles the line between science of AI and science of astronomy. Photometric redshift estimation is a well-established subfield of astronomy. Prior works show that computer vision models typically outperform catalog-based models, but these models face additional complexities when incorporating images from more than one instrument or sensor. In this report, we detail our progress creating Mantis Shrimp, a multi-survey computer vision model for photometric redshift estimation that fuses ultra-violet (GALEX), optical (PanSTARRS), and infrared (UnWISE) imagery. We use deep learning interpretability diagnostics to measure how the model leverages information from the different inputs. We reason about the behavior of the CNNs from the interpretability metrics, specifically framing the result in terms of physically-grounded knowledge of galaxy properties.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑