Modern time-domain surveys continuously monitor large swaths of the sky to look for astronomical variability. Astrophysical discovery in such data sets is complicated by the fact that detections of real transient and variable sources are highly outnumbered by bogus detections caused by imperfect subtractions, atmospheric effects and detector artefacts. In this work we present a machine learning (ML) framework for discovery of variability in time-domain imaging surveys. Our ML methods provide probabilistic statements, in near real time, about the degree to which each newly observed source is astrophysically relevant source of variable brightness. We provide details about each of the analysis steps involved, including compilation of the training and testing sets, construction of descriptive image-based and contextual features, and optimization of the feature subset and model tuning parameters. Using a validation set of nearly 30,000 objects from the Palomar Transient Factory, we demonstrate a missed detection rate of at most 7.7% at our chosen false-positive rate of 1% for an optimized ML classifier of 23 features, selected to avoid feature correlation and over-fitting from an initial library of 42 attributes. Importantly, we show that our classification methodology is insensitive to mis-labelled training data up to a contamination of nearly 10%, making it easier to compile sufficient training sets for accurate performance in future surveys. This ML framework, if so adopted, should enable the maximization of scientific gain from future synoptic survey and enable fast follow-up decisions on the vast amounts of streaming data produced by such experiments.
KMTNet Synoptic Survey of Southern Sky III: The First Data Release
Seoul National University（首尔大学）；Daegu National Science Museum（大邱国立科学馆）；University of Cape Town（开普敦大学）；South African Astronomical Observatory（南非天文台）；University of the Free State（自由邦大学）；Swinburne University of Technology（斯威本科技大学）；ARC Centre of Excellence for Gravitational Wave Discovery (OzGrav)（ARC引力波发现卓越中心（OzGrav））；Korea Astronomy and Space Science Institute（韩国天文学与空间科学研究所）
Toward automated detection of light echoes in synoptic surveys: considerations on the application of the Deep Convolutional Neural Networks
光回波（LE）是星际尘埃对天体物理瞬变的反射。它们是令人着迷的天文现象，使人们能够研究散射的尘埃以及原始的瞬变。然而，LE是罕见的，并且极难检测，因为它们看起来是微弱的、扩散的、随时间变化的特征。LE的检测仍然主要依赖于人类对图像的检查，这是一种在大型天气调查时代不可行的方法。维拉C。Rubin Observatory Legacy Survey of Space and Time，LSST，将以高空间分辨率，精致的图像质量和超过数万平方度的天空生成前所未有的天文成像数据：这是LE的理想调查。然而，Rubin数据处理管道针对点源的检测进行了优化，并且将完全错过LE。在过去的几年里，人工智能（AI）目标检测框架已经实现并超越了实时、人类水平的性能。在这项工作中，我们准备了一个来自ATLAS望远镜的数据集，并测试了一个流行的AI对象检测框架，即计算机视觉社区开发的You Only Look Once或YOLO，以展示AI在天文图像中检测LE的潜力。我们发现，即使是在大小和质量有限的数据集上，AI框架也可以达到人类水平的性能。我们探索并强调了挑战，包括类别不平衡和标签不完整性，并为高通量天文调查中LE的自动检测和研究建立端到端管道所需的工作制定了路线图。
The RATT PARROT: serendipitous discovery of a peculiarly scintillating pulsar in MeerKAT imaging observations of the Great Saturn-Jupiter Conjunction of 2020. I. Dynamic imaging and data analysis
VASTER: The ASKAP real-time fast-imaging pipeline -- overview and discovery of two long period transients
Swinburne University of Technology（斯威本科技大学）；The University of Sydney（悉尼大学）；CSIRO Space and Astronomy（澳大利亚联邦科学与工业研究组织空间与天文部）；University of Wisconsin-Milwaukee（威斯康星大学密尔沃基分校）；Curtin University（科廷大学）
SCExAO/CHARIS and Gaia Direct Imaging and Astrometric Discovery of a Superjovian Planet 3--4 lambda/D from the Accelerating Star HIP 54515
University of Texas at San Antonio（圣安东尼奥德克萨斯大学）；University of Michigan（密歇根大学）；University of California-Santa Cruz（加州大学圣克鲁兹分校）；University of Hawai'i（夏威夷大学）；Astrobiology Center, NINS（国立天文台天体生物学中心（NINS））；National Astronomical Observatory of Japan（日本国立天文台）；University of Notre Dame（圣母大学）；The University of Tokyo（东京大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；Space Telescope Science Institute（太空望远镜科学研究所）；NASA-Goddard Space Flight Center（美国国家航空航天局戈达德太空飞行中心）；Stockholm University（斯德哥尔摩大学）；Princeton University（普林斯顿大学）；National Research Council, Herzberg Astronomy and Astrophysics（加拿大国家研究委员会赫茨伯格天文学与天体物理学研究所）；Jet Propulsion Laboratory, California Institute of Technology（加州理工学院喷气推进实验室）；The Graduate University for Advanced Studies, SOKENDAI（综合研究大学院大学（SOKENDAI））；Hokkaido University（北海道大学）；New Mexico State University（新墨西哥州立大学）；Stony Brook University（石溪大学）；University of Hawai’i at Manoa（夏威夷大学马诺阿分校）；California State University Northridge（北岭加州州立大学）
Enabling Early Transient Discovery in LSST via Difference Imaging with DECam
University of California, Santa Barbara（加州大学圣塔芭芭拉分校）；Rutgers, the State University of New Jersey（新泽西州立罗格斯大学）；Massachusetts Institute of Technology（麻省理工学院）；Steward Observatory, University of Arizona（亚利桑那大学斯图尔德天文台）；University of Oxford（牛津大学）；SLAC National Accelerator Laboratory（斯坦福直线加速器中心）；Kavli Institute for Particle Astrophysics and Cosmology, SLAC National Accelerator Laboratory（SLAC国家加速器实验室卡弗里粒子天体物理与宇宙学研究所）；Smithsonian Astrophysical Observatory（史密森尼天体物理台）；University of Washington（华盛顿大学）；Yonsei University（延世大学）；Princeton University（普林斯顿大学）；Brookhaven National Laboratory（布鲁克海文国家实验室）；Institute for Data-intensive Research in Astrophysics and Cosmology, University of Washington（华盛顿大学天体物理学与宇宙学数据密集型研究所）；Columbia University（哥伦比亚大学）；California Institute of Technology（加州理工学院）；Duke University（杜克大学）；Purdue University（普渡大学）
POLISH'ing the Sky: Wide-Field and High-Dynamic Range Interferometric Image Reconstruction with Application to Strong Lens Discovery
Department of Computing and Mathematical Sciences, California Institute of Technology (Caltech), Pasadena, CA, USA（计算与数学科学系，加州理工学院（Caltech），帕萨迪纳，CA，美国）；Center for Astrophysics ∣ \mid Harvard & Smithsonian, Cambridge, MA, USA（天体物理中心，哈佛与史密松尼天体物理研究中心，剑桥，MA，美国）
