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 2509.22018astro-ph.COastro-ph.IMcs.LG
收藏

利用深度学习探索早期宇宙

Exploring the Early Universe with Deep Learning

Haute Ecole Arc Ing\'enierie, University of Applied Sciences \& Arts Western Switzerland (HES-SO), Saint-Imier, Switzerland Département de Physique Théorique;Centre Universitaire d’Informatique, Université de Genève, Genève, Switzerland Nordita, KTH Royal Institute of Technology;Stockholm University, Hannes Alf\'vens v\"ag 12, SE-106 91 Stockholm, Sweden Institute for Particle Physics \& Astrophysics, ETH Zurich, Wolfgang-Pauli-Str 27, 8093 Zurich, Switzerland Laboratoire d’Astrophysique, Ecole Polytechnique Federale de Lausanne EPFL, Observatoire de Sauverny, Versoix 1290, Switzerland Centre for Artificial Intelligence, ZHAW Zurich University of Applied Sciences, Technikumstrasse 71, 8400 Winterthur, Switzerland Institute for Data Science, FHNW University of Applied Sciences \& Arts Northwestern Switzerland, Bahnhofstrasse 6, Windisch, 5210, Switzerland Department of Space, Earth;Environment, Chalmers University of Technology, Onsala Space Observatory, SE-43992 Onsala, Sweden
Emmanuel de Salis, Massimo De Santis, Davide Piras, Sambit K. Giri, Michele Bianco, Nicolas Cerardi, Philipp Denzel, Merve Selcuk-Simsek, Kelley M. Hess, M. Car… 展开作者

Emmanuel de Salis, Massimo De Santis, Davide Piras, Sambit K. Giri, Michele Bianco, Nicolas Cerardi, Philipp Denzel, Merve Selcuk-Simsek, Kelley M. Hess, M. Carmen Toribio, Franz Kirsten, Hatem Ghorbel

AI总结:

本文提出利用深度学习从SKAO预期的氢信号二维功率谱中提取信息,预测宇宙再电离历史,实验表明专用机器学习算法平均R²超过0.95,显著提升早期宇宙结构形成的推断精度。

摘要:

氢是宇宙中最丰富的元素。第一代恒星和星系产生的光子电离了氢气,驱动了一个被称为再电离纪元(EoR)的宇宙学事件。即将建成的平方公里阵列天文台(SKAO)将绘制这一时期中性氢的分布图,有助于研究这些第一代天体的性质。提取天体物理信息将具有挑战性,因为SKAO将产生海量数据,其中氢信号会受到不期望的前景污染和仪器系统效应的干扰。为了解决这一问题,我们开发了最新的深度学习技术,从SKAO预期的氢信号二维功率谱中提取信息。我们将一系列神经网络模型应用于这些测量,并量化它们预测宇宙氢再电离历史的能力,该历史与早期光子源数量的增加和效率的提高相关。我们表明,早期宇宙的研究受益于现代深度学习技术。特别是,我们证明了专用的机器学习算法在恢复再电离历史方面平均可以达到超过0.95的R²分数。这使得对早期宇宙中结构形成的宇宙学和天体物理学推断能够准确而精确地进行。

英文摘要:

Hydrogen is the most abundant element in our Universe. The first generation of stars and galaxies produced photons that ionized hydrogen gas, driving a cosmological event known as the Epoch of Reionization (EoR). The upcoming Square Kilometre Array Observatory (SKAO) will map the distribution of neutral hydrogen during this era, aiding in the study of the properties of these first-generation objects. Extracting astrophysical information will be challenging, as SKAO will produce a tremendous amount of data where the hydrogen signal will be contaminated with undesired foreground contamination and instrumental systematics. To address this, we develop the latest deep learning techniques to extract information from the 2D power spectra of the hydrogen signal expected from SKAO. We apply a series of neural network models to these measurements and quantify their ability to predict the history of cosmic hydrogen reionization, which is connected to the increasing number and efficiency of early photon sources. We show that the study of the early Universe benefits from modern deep learning technology. In particular, we demonstrate that dedicated machine learning algorithms can achieve more than a $0.95$ $R^2$ score on average in recovering the reionization history. This enables accurate and precise cosmological and astrophysical inference of structure formation in the early Universe.

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。

↑