NASA的Kepler Space Telescope在寻找银河系中系外行星存在的任务中发挥了关键作用。这一搜索得到了计算数据分析的支持,旨在从Kepler望远镜接收到的信号中识别系外行星。本文考虑在Kepler太空望远镜及其扩展任务K2的数据上,利用残差网络对现有系外行星识别工作进行扩展。本文旨在探索深度学习算法如何帮助在一种数据量较少的情况下和另一种数据种类更广泛的情况下对系外行星的存在进行分类。除了基于标准CNN的方法外,我们提出了一种Siamese架构,该架构特别适用于解决低数据场景下的分类问题。CNN和ResNet算法在三类分类中达到了68%的平均准确率,在二类分类中达到了86%的准确率。然而,对于三类和二类分类,Siamese算法均达到了99%的准确率。
英文摘要:
NASA's Kepler Space Telescope has been instrumental in the task of finding the presence of exoplanets in our galaxy. This search has been supported by computational data analysis to identify exoplanets from the signals received by the Kepler telescope. In this paper, we consider building upon some existing work on exoplanet identification using residual networks for the data of the Kepler space telescope and its extended mission K2. This paper aims to explore how deep learning algorithms can help in classifying the presence of exoplanets with less amount of data in one case and a more extensive variety of data in another. In addition to the standard CNN-based method, we propose a Siamese architecture that is particularly useful in addressing classification in a low-data scenario. The CNN and ResNet algorithms achieved an average accuracy of 68% for three classes and 86% for two-class classification. However, for both the three and two classes, the Siamese algorithm achieved 99% accuracy.
Automated identification of transiting exoplanet candidates in NASA Transiting Exoplanets Survey Satellite (TESS) data with machine learning methods
Catholic University of America(美国天主教大学);Tel Aviv University(特拉维夫大学);ThetaRay(ThetaRay公司)
一种使用机器学习(ML)方法的新型人工智能(AI)技术结合了由ThetaRay,Inc.应用于NASA的凌日系外行星勘测卫星(TESS)数据集,以识别系外行星候选者。AI/ML ThetaRay系统最初使用开普勒系外行星数据进行训练,并在应用于TESS数据之前使用确认的系外行星进行验证。数据的现有和新特征基于各种观测参数,通过采用半监督和无监督机器学习技术构建并用于AI/ML分析。通过将ThetaRay系统应用于TESS任务产生的10,803个阈值交叉事件(TCEs)的光变曲线,该算法从Mikulski Archive for Space Telescopes获得,该算法产生了约50个目标以供进一步分析,我们通过进一步的人工审查发现了三个新的系外行星候选者。这项研究首次展示了将特定的基于AI/ML的方法成功应用于大型天体物理数据集,以快速自动分类TCE。
Exoplanet Transit Candidate Identification in TESS Full-Frame Images via a Transformer-Based Algorithm
Pontificia Universidad Católica de Chile(智利天主教大学);The Catholic University of America(美国天主教大学);NASA Goddard Space Flight Center(美国国家航空航天局戈达德太空飞行中心);Universidad Adolfo Ibáñez(阿道夫·伊巴涅斯大学);Millennium Institute for Astrophysics(千禧年天体物理学研究所);University of Maryland(马里兰大学);Oak Ridge Associated Universities(橡树岭联合大学);KU Leuven(鲁汶大学)
Exoplanet Classification through Vision Transformers with Temporal Image Analysis
Department of Mathematics and Computing, Indian Institute of Technology (Indian School of Mines), Dhanbad 826004, India(数学与计算系,印度理工学院(印度矿山学院),达翰巴德);Department of Electronics Engineering, Indian Institute of Technology (Indian School of Mines), Dhanbad 826004, India(电子工程系,印度理工学院(印度矿山学院),达翰巴德);Indian Institute of Astrophysics, Koramangala 2nd Block, Bangalore 560034, India(印度天体物理研究所,班加罗尔);Pondicherry University, R.V. Nagar, Kalapet, 605014, Puducherry, India(英拉大学)
The Identification of CS2 and Evidence for Carbon-Sulfur Chemical Coupling in a Warm Giant Exoplanet Atmosphere
University of Wisconsin–Madison(威斯康星大学麦迪逊分校);Arizona State University(亚利桑那州立大学);Space Telescope Science Institute(太空望远镜科学研究所);Schmidt Sciences(施密特科学);California Institute of Technology(加州理工学院);University of California Santa Cruz(加州大学圣克鲁兹分校);Université Côte d’Azur(蔚蓝海岸大学)
Analysis of Habitability and Stellar Habitable Zones from Observed Exoplanets
Jet Propulsion Laboratory, California Institute of Technology(加州理工学院喷气推进实验室);Lakeside School(湖畔学校);Pacific Academy(太平洋学院);Burke Institute for Theoretical Physics, California Institute of Technology(加州理工学院伯克理论物理研究所)
Vetting Ambiguous Planetary Candidates (APC) in TESS Data: Insights from LATTE Package and Detection of Nine Potential Exoplanet Candidates
Pakistan Space and Upper Atmosphere Research Commission (SUPARCO)(巴基斯坦空间与高层大气研究委员会);Institute of Frontier and Interdisciplinary Science, Shandong University(山东大学前沿交叉科学研究院)
Identifying Exoplanets with Deep Learning. V. Improved Light Curve Classification for TESS Full Frame Image Observations
Massachusetts Institute of Technology(麻省理工学院);Google(谷歌);University of Southern Queensland(南昆士兰大学);Princeton University(普林斯顿大学)
TESS任务产生了大量的时间序列数据,其中只有一小部分包含可检测的系外行星过境信号。神经网络等深度学习技术已被证明可以有效地将有希望的天体物理学日食候选者与其他现象(如恒星变化和系统性仪器效应)区分开来,并以有效,无偏见和可持续的方式进行。本文介绍了一个高质量的数据集,其中包含来自主要任务和第一次扩展任务全帧图像的光变曲线以及通过Box Least Squares检测到的周期信号(Kovács等人,2002; Hartman 2012)。该数据集使用彻底的手动审查过程进行管理,然后用于训练称为Astronet-Triage-v2的神经网络。在我们的测试集上,对于过境/日食事件,我们实现了99.6%的召回率(所有带有阳性标签的数据的真阳性),精确度为75.7%(所有预测阳性的真阳性)。由于我们90%的训练数据来自主要任务,我们还测试了我们对第一次扩展任务数据的概括能力。在这里,我们发现精确度-召回率曲线下面积为0.965,比Astronet-Triage提高了4%(Yu et al. 2019)。在截至2022年4月的TESS感兴趣对象(TOI)目录(行星和行星候选者的入围名单)中,Astronet-Triage-v2能够恢复4140个TOI中的3577个,而Astronet-Triage仅以同等精度恢复3349个目标。换句话说,升级到Astronet-Triage-v2可以帮助至少…
Postulating Exoplanetary Habitability via a Novel Anomaly Detection Method
BITS Pilani, K. K. Birla Goa Campus(印度博拉理工学院果阿分校);CSIS and APPCAIR, BITS Pilani, K. K. Birla Goa Campus(印度博拉理工学院果阿分校CSIS与APPCAIR中心);Indian Institute of Astrophysics(印度天体物理研究所)
Machine-learning clustering of close-in exoplanet populations: links to pebble accretion
Center for Star and Planet Formation(星系与行星形成中心);Globe Institute(全球研究所);University of Copenhagen(哥本哈根大学);Lund Observatory(隆德天文台);Department of Physics(物理系);Lund University(隆德大学);Graduate Institute of Astronomy(天文研究所);National Central University(国立中央大学)
Machine Learning and Deep Learning for Exoplanet Detection and Atmospheric Characterization with JWST and the Upcoming Ariel Mission
Department of Industrial Physics, Enugu State University of Science and Technology(恩古州科技大学工业物理系);Department of Physics, Dennis Osadebay University(德尼斯·奥萨德贝大学物理系)
YOUNG Star detrending for Transiting Exoplanet Recovery (YOUNGSTER) II: Using Self-Organising Maps to explore young star variability in Sectors 1-13 of TESS data
Dept. of Physics, University of Warwick(华威大学物理系);Centre for Exoplanets and Habitability, University of Warwick(华威大学系外行星与宜居性中心)
Classification and Nomenclature of Planets in the Mass-Radius Plane
St. Joseph’s University(圣约瑟夫大学);Christ (Deemed to be University)(基督(认定)大学);M. P. Birla Institute of Fundamental Research(M.P.比尔拉基础研究所);LUNEX EuroMoonMars EuroSpaceHub(LUNEX 欧空月球火星欧洲空间枢纽);University of Latvia(拉脱维亚大学);NoRCEL Institute(诺塞尔研究所)
Bayesian Deep Learning for Exoplanet Atmospheric Retrieval
King’s College London(伦敦国王学院);University of Central Florida(中佛罗里达大学);University of Pittsburgh(匹兹堡大学);Luxembourg Institute of Science and Technology(卢森堡科学技术研究院);University of Oxford(牛津大学);University of Bern(伯尔尼大学);Google Cloud(谷歌云);NASA Astrobiology Institute(美国国家航空航天局天体生物学研究所);NASA Goddard Space Flight Center(美国国家航空航天局戈达德太空飞行中心)
Machine Learning for Exoplanet Detection in High-Contrast Spectroscopy: Revealing Exoplanets by Leveraging Hidden Molecular Signatures in Cross-Correlated Spectra with Convolutional Neural Networks
Institute for Particle Physics and Astrophysics, ETH Zürich(苏黎世联邦理工学院粒子物理与天体物理研究所);Seminar für Statistik, ETH Zürich(苏黎世联邦理工学院统计研讨会);Department of Astronomy, University of Michigan(密歇根大学天文系);Département d’Astronomie, Université de Genève(日内瓦大学天文系);STAR Institute, University of Liège(列日大学STAR研究所)
本文提出 MLCCS,用感知机和一维卷积神经网络分析互相关光谱中的隐藏分子信号,在 SINFONI K 波段模拟数据上显著提升暗弱系外行星探测灵敏度。
Inclusion of sulfur chemistry in a validated C/H/O/N chemical network: identification of key C/S coupling pathways
Université Paris Cité and Univ Paris Est Creteil, CNRS, LISA(巴黎西岱大学与巴黎东克雷泰伊大学,法国国家科学研究中心,LISA);Astrophysics Group, University of Exeter(埃克塞特大学天体物理学组);Université de Lorraine, CNRS, LRGP(洛林大学,法国国家科学研究中心,LRGP);Department of Physics and Astronomy, University College London(伦敦大学学院物理与天文系)
Challenges in the detection of gases in exoplanet atmospheres
Arizona State University(亚利桑那州立大学);University of Maryland(马里兰大学);Space Science Institute(空间科学研究所);Lawrence Livermore National Laboratory(劳伦斯利弗莫尔国家实验室);University of California Santa Cruz(加州大学圣克鲁兹分校);Johns Hopkins University(约翰斯·霍普金斯大学);Michigan State University(密歇根州立大学);University of California, Los Angeles(加州大学洛杉矶分校);Université de Montréal(蒙特利尔大学)