arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.22543astro-ph.GAastro-ph.CO

深度学习晕定义器:用于星系群和星系团晕质量与气体分数推断的多模态框架

The Deep Learning Halo Definer: A Multimodal Framework for Halo Mass and Gas Fraction Inference on Galaxy Groups and Clusters

Caleb Ogle, Kalvyn N. Poncelet Adams, Benjamin D. Oppenheimer, Naomi Gluck, Matthew Ho, Daisuke Nagai, Joseph N. Burchett, Mohammadreza Ayromlou

首次发表
浏览论文内容

中文总结 AI 辅助

针对星系群和星系团尺度上暗物质晕质量与气体分数推断的挑战,提出多模态深度学习框架DLHD,结合深度集合与CNN处理星系目录和X射线数据,在模拟数据上显著降低RMSE,优于单一模态方法。

中文摘要 AI 辅助

在星系群尺度上,准确推断暗物质晕属性(如总晕质量 M_{halo} 和气体分数 f_{gas})仍然特别具有挑战性,因为该尺度下成员数量少、势阱浅,以及活动星系核反馈驱动的重子抛射会引入显著的观测离散度。随着大规模巡天开始提供前所未有的多波长数据,迫切需要能够联合利用多种可观测量的方法来克服这些不确定性。我们提出了深度学习晕定义器(DLHD),这是一个多模态深度学习框架,通过结合深度集合(Deep Sets)和卷积神经网络(CNN),同时处理星系目录和X射线成像,以改进星系群和星系团的 M_{halo} 和 f_{gas} 估计。利用源自 IllustrisTNG300 流体动力学模拟的模拟数据集,我们证明 DLHD 的性能优于其每个单独组成的网络,在 M_{halo} 的均方根误差(RMSE)上,相比 Deep Sets 提高了1.9倍,相比单独使用 CNN 提高了1.3倍。对于 R_{200c} 内包含的气体分数,DLHD 相对于 Deep Sets 将 RMSE 降低了2.0倍,相对于单独使用 CNN 降低了1.1倍,并且在其他孔径上也保持一致改进。这些结果突显了一种利用单模态方法无法获取的多波段信息的新能力,使 DLHD 成为下一代巡天分析的有力工具。

英文摘要

Accurately inferring dark matter halo properties like the total halo mass (M_{halo}) and gas fractions (f_{gas}) remains particularly challenging at group scales, where low member counts, shallow potential wells, and AGN feedback-driven baryon expulsion introduce significant observational scatter. As large-scale surveys begin to provide unprecedented multi-wavelength data, there is a pressing need for methods that can jointly leverage diverse observables to overcome these uncertainties. We introduce the Deep Learning Halo Definer (DLHD), a multimodal deep learning framework that simultaneously processes galaxy catalogues and X-ray imaging through the combination of a Deep Sets and a Convolutional Neural Network (CNN) to improve M_{halo} and f_{gas} estimation for galaxy groups and clusters. Using mock datasets derived from the IllustrisTNG300 hydrodynamic simulation, we demonstrate that the DLHD outperforms each of its component networks individually, achieving RMSE improvements in M_{halo} of 1.9x over Deep Sets and 1.3x over the CNN alone. For the gas fraction enclosed within R_{200c}, DLHD reduces RMSE by 2.0x relative to Deep Sets and 1.1x relative to the CNN alone, with consistent improvements across other apertures. These results highlight a novel ability to leverage multi-band information inaccessible to single-modality methods, positioning DLHD as a promising tool for next-generation survey analyses.

发表机构

  • University of Wisconsin-Milwaukee(密尔沃基威斯康星大学)
  • University of California, Los Angeles(加州大学洛杉矶分校)
  • University of Colorado(科罗拉多大学)
  • Yale University(耶鲁大学)
  • Sorbonne Université(索邦大学)
  • Columbia University(哥伦比亚大学)
  • New Mexico State University(新墨西哥州立大学)
  • Argelander-Institut für Astronomie(阿尔兰德天体物理研究所)

机构由 AI 辅助整理,请以论文原文为准。

补充信息

↑