The next generation of observatories will facilitate the discovery of new types of astrophysical transients. The detection of such phenomena, whose characteristics are presently poorly constrained, will hinge on the ability to perform blind searches. We present a new algorithm for this purpose, based on deep learning. We incorporate two approaches, utilising anomaly detection and classification techniques. The first is model-independent, avoiding the use of background modelling and instrument simulations. The second method enables targeted searches, relying on generic spectral and temporal patterns as input. We compare our methodology with the existing approach to serendipitous detection of gamma-ray transients. We use our framework to derive the detection prospects of low-luminosity gamma-ray bursts with the upcoming Cherenkov Telescope Array. Our method is an unbiased, data-driven approach for multiwavelength and multi-messenger transient detection.
The Antares neutrino telescope, operating at 2.5 km depth in the Mediterranean Sea, 40 km off the Toulon shore, represents the world's largest operational underwater neutrino telescope, optimized for the detection of Cerenkov light produced by neutrino-induced muons. The main goal of Antares is the search of high energy neutrinos from astrophysical point or transient sources. Antares is taking data in its full 12 lines configuration since May 2008: in this paper we collect the 16 contributions by the ANTARES collaboration that were submitted to the 31th International Cosmic Ray Conference ICRC…
Cosmic-Ray Physics in the PeV to EeV Energy Range with the IceCube-Gen2 Surface Array (ICRC 2025)
Bartol Research Institute, Department of Physics and Astronomy, University of Delaware（特拉华大学巴托尔物理与天文学研究所）；Institute for Astroparticle Physics, Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院高能粒子物理研究所）
This conference proceedings is a write-up of the Gamma-ray Indirect rapporteur talk given at the 37th International Cosmic Ray Conference (ICRC 2021). In contrast to previous ICRCs, this years edition was held in a fully virtual format, with dedicated discussion sessions organised around specific scientific themes. Many of these topics span the two categories of Gamma-ray Indirect (GAI) and Gamma-ray Direct (GAD), observations of gamma-rays by ground-based and space-based facilities respectively. To cover this organisation by topic in a coherent manner, this GAI rapporteur contribution focuses…
IceCube-Gen2, the extension of the IceCube Neutrino Observatory, will feature three main components: an optical array in the deep ice, a large-scale radio array in the shallow ice and firn, and a surface detector above the optical array. Thus, IceCube-Gen2 will not only be an excellent detector for PeV neutrinos, but also constitutes a unique setup for the measurement of cosmic-ray air showers, where the electromagnetic component and low-energy muons are measured at the surface and high-energy muons are measured in the ice. As for ongoing enhancement of IceCube's current surface array, IceTop,…
We report the results of a low frequency radio variability and slow transient search using archival observations from the Very Long Array. We selected six 325 MHz radio observations from the spring of 2006, each centered on the Spitzer-Space-Telescope Wide-area Infrared Extragalactic Survey (SWIRE) Deep Field: 1046+59. Observations were spaced between 1 day to 3 months, with a typical single-epoch peak flux sensitivity below 0.2 \mjb near the field pointing center. We describe the observation parameters, data post-processing, and search methodology used to identify variable and transient emiss…
[Abridged] We present a search for fast optical transients (~0.5 hr-1 day) using repeated observations of the Pan-STARRS1 Medium-Deep Survey (PS1/MDS) fields. Our search takes advantage of the consecutive g/r-band observations (16.5 min in each filter), by requiring detections in both bands, with non-detections on preceding and subsequent nights. We identify 19 transients brighter than 22.5 AB mag (S/N>10). Of these, 11 events exhibit quiescent counterparts in the deep PS1/MDS templates that we identify as M4-M9 dwarfs. The remaining 8 transients exhibit a range of properties indicative of mai…
The next generation of observatories will facilitate the discovery of new types of astrophysical transients. The detection of such phenomena, whose characteristics are presently poorly constrained, will hinge on the ability to perform blind searches. We present a new algorithm for this purpose, based on deep learning. We incorporate two approaches, utilising anomaly detection and classification techniques. The first is model-independent, avoiding the use of background modelling and instrument simulations. The second method enables targeted searches, relying on generic spectral and temporal pat…
It is assumed that accreting neutron stars (NSs) in LMXBs are heated due to the compression of the existing crust by the accreted matter which gives rise to nuclear reactions in the crust. It has been shown that most of the energy is released deep in the crust by pycnonuclear reactions involving low-Z elements. We discuss if NSs in very-faint X-ray transients (VFXTs; those which have peak X-ray luminosities < 1E36 erg/s) can be used to test this model. Unfortunately we cannot conclusively answer this because of the large uncertainties in our estimates of the accretion rate history of those VFX…
Recent studies have shown possible connections between highly magnetized neutron stars ("magnetars"), whose X-ray emission is too bright to be powered by rotational energy, and ordinary radio pulsars. In addition to the magnetar SGR J1745-2900, one of the radio pulsars in the Galactic centre (GC) region, PSR J1746-2850, had timing properties implying a large magnetic field strength and young age, as well as a flat spectrum. All characteristics are similar to those of rare, transient, radio-loud magnetars. Using several deep non-detections from the literature and two new detections, we show tha…
Ground-based $γ$-ray observatories, such as the VERITAS array of imaging atmospheric Cherenkov telescopes, provide insight into very-high-energy (VHE, $\mathrm{E}>100\,\mathrm{GeV}$) astrophysical transient events. Examples include the evaporation of primordial black holes, gamma-ray bursts and flaring blazars. Identifying such events with a serendipitous location and time of occurrence is difficult. Thus, employing a robust search method becomes crucial. An implementation of a transient detection method based on deep-learning techniques for VERITAS will be presented. This data-driven approach…
Columbia University（哥伦比亚大学）；Princeton University（普林斯顿大学）；Center for Astrophysics | Harvard & Smithsonian（哈佛与史密森尼天体物理中心）；Flatiron Institute（熨斗研究所）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
There is a shortage of multi-wavelength and spectroscopic followup capabilities given the number of transient and variable astrophysical events discovered through wide-field, optical surveys such as the upcoming Vera C. Rubin Observatory. From the haystack of potential science targets, astronomers must allocate scarce resources to study a selection of needles in real time. Here we present a variational recurrent autoencoder neural network to encode simulated Rubin Observatory extragalactic transient events using 1% of the PLAsTiCC dataset to train the autoencoder. Our unsupervised method uniqu…
We report on an analysis of the Extended Chandra Deep Field South (E-CDFS) region using archival data from the Very Large Array, with the goal of studying radio variability and transients at the sub-mJy level. The 49 epochs of E-CDFS observations at 1.4 GHz sample timescales from one day to 3 months. We find that only a fraction (1%) of unresolved radio sources above 40 uJy are variable at the 4-sigma level. There is no evidence that the fractional variability changes along with the known transition of radio source populations below one milliJansky. Optical identifications of the sources show …
We present a deep neural network Real/Bogus classifier that improves classification performance in the Tomo-e Gozen transient survey by handling label errors in the training data. In the wide-field, high-frequency transient survey with Tomo-e Gozen, the performance of conventional convolutional neural network classifier is not sufficient as about $10^6$ bogus detections appear every night. In need of a better classifier, we have developed a new two-stage training method. In this training method, label errors in the training data are first detected by normal supervised learning classification, …
This list of contributions to the 37th International Cosmic Ray Conference in Berlin, Germany (12-23 July 2021) summarizes the latest results from the IceCube Neutrino Observatory. IceCube, completed 10 years ago at the geographic South Pole, comprises a surface detector designed to observe cosmic ray air showers, a cubic-kilometer array of optical sensors deployed deep in the ice sheet to observe TeV-PeV neutrinos, and a 15 Megaton deep-ice subdetector sensitive to >10 GeV neutrinos. Data from IceCube are used to investigate a broad set of key questions in physics and astrophysics, such as th…
We report on the results from a large observational campaign on the bare Seyfert galaxy Ark 120, jointly carried out in 2014 with XMM-Newton, Chandra, and NuSTAR. The fortunate line of sight to this source, devoid of any significant absorbing material, provides an incomparably clean view to the nuclear regions of an active galaxy. Here we focus on the analysis of the iron fluorescence features, which form a composite emission pattern in the 6$-$7 keV band. The prominent K$α$ line from neutral iron at 6.4 keV is resolved in the Chandra High-Energy Transmission Grating spectrum to a full-width a…
LUNCH: A Lightweight Unified Deep-Learning Framework for General Transients Classification in High-Energy Time-Domain Astronomy
College of Electronic and Information Engineering, Tongji University（同济大学电子与信息工程学院）；State Key Laboratory of Particle Astrophysics, Institute of High Energy Physics, Chinese Academy of Sciences（中国科学院高能物理研究所粒子天体物理国家重点实验室）
University of Cambridge（剑桥大学）；Space Telescope Science Institute（太空望远镜科学研究所）；Kavli Institute for Cosmology（卡夫利宇宙学研究所）；Stockholm University（斯德哥尔摩大学）；University of Toronto（多伦多大学）
We present RAPID (Real-time Automated Photometric IDentification), a novel time-series classification tool capable of automatically identifying transients from within a day of the initial alert, to the full lifetime of a light curve. Using a deep recurrent neural network with Gated Recurrent Units (GRUs), we present the first method specifically designed to provide early classifications of astronomical time-series data, typing 12 different transient classes. Our classifier can process light curves with any phase coverage, and it does not rely on deriving computationally expensive features from…
Recent works have discovered two fast ($\approx 10$ ks) extragalactic X-ray transients in the Chandra Deep Field-South (CDF-S XT1 and XT2). These findings suggest that a large population of similar extragalactic transients might exist in archival X-ray observations. We develop a method that can effectively detect such transients in a single Chandra exposure, and systematically apply it to Chandra surveys of CDF-S, CDF-N, DEEP2, UDS, COSMOS, and E-CDF-S, totaling 19~Ms of exposure. We find 13 transient candidates, including CDF-S XT1 and XT2. With the aid of available excellent multiwavelength …
The observed light curves and other properties of the two extragalactic fast x-ray transients, CDF-S XT1 and CDF-S XT2, which were discovered recently in archival data of the Chandra Deep Field-South (CDF-S) observations, indicate that they belong to two different populations of X-ray transients. XT1 seems to be an x-ray flash (XRF), i.e., a narrowly beamed long duration gamma ray burst viewed from far off-axis while XT2 seems to be a nebular emission powered by a newly born millisecond pulsar in a neutron stars binary merger.
Searches for optical transients are usually performed with a cadence of days to weeks, optimised for supernova discovery. The optical fast transient sky is still largely unexplored, with only a few surveys to date having placed meaningful constraints on the detection of extragalactic transients evolving at sub-hour timescales. Here, we present the results of deep searches for dim, minute-timescale extragalactic fast transients using the Dark Energy Camera, a core facility of our all-wavelength and all-messenger Deeper, Wider, Faster programme. We used continuous 20s exposures to systematically…
Gamma-ray astronomy is a branch whose potential has not yet been fully exploited. The observations of elemental and isotopic abundances in supernova (SN) explosions are key probes not only of the stellar structure and evolution but also for understanding the physics that makes Type-Ia SNe as standard candles for the study of the Universe expansion properties. In spite of its crucial role, nuclear astrophysics remains a poorly explored field mainly for the typical emission lines intensity which are vanishing small and requires very high sensitivities of the telescopes. Furthermore, in spite tha…
SN 1954J in NGC 2403 and SN 1961V in NGC 1058 were two luminous transients whose definitive classification as either non-terminal eruptions or supernovae remains elusive. A critical question is whether a surviving star can be significantly obscured by dust formed from material ejected during the transient. We use three lines of argument to show that the candidate surviving stars are not significantly optically extincted ($τ\lesssim 1$) by dust formed in the transients. First, we use SED fits to new HST optical and near-IR photometry. Second, neither source is becoming brighter as required by a…
The primary challenge in the study of explosive astrophysical transients is their detection and characterisation using multiple messengers. For this purpose, we have developed a new data-driven discovery framework, based on deep learning. We demonstrate its use for searches involving neutrinos, optical supernovae, and gamma rays. We show that we can match or substantially improve upon the performance of state-of-the-art techniques, while significantly minimising the dependence on modelling and on instrument characterisation. Particularly, our approach is intended for near- and real-time analys…
The Deeper, Wider, Faster programme's first DECam optical data release
Swinburne University of Technology（斯威本科技大学）；University of Sydney（悉尼大学）；Macquarie University（麦考瑞大学）；University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）
NEXUS: Transient Searches and First Results from Year One Observations
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Pennsylvania（宾夕法尼亚大学）；Carnegie Mellon University（卡内基梅隆大学）；Space Telescope Science Institute（太空望远镜科学研究所）；UC San Diego（加州大学圣地亚哥分校）；University of Arizona（亚利桑那大学）；Peking University（北京大学）
We present results from a total of 459 repeated 3.1 GHz radio continuum observations (of which 379 were used in a search for transient sources) of the ELAIS-N1, Coma, Lockman Hole, and NOAO Deep Wide Field Survey fields as part of the Pi GHz Sky Survey (PiGSS). The observations were taken approximately once per day between 2009 May and 2011 April. Each image covers 11.8 square degrees and has 100 arcsecond FWHM resolution. Deep images for each of the four fields have rms noise between 180 and 310 uJy and the corresponding catalogs contain ~200 sources in each field. Typically 40 - 50 of these …
ULTIMATE-Subaru (Ultra-wide Laser Tomographic Imager and MOS with AO for Transcendent Exploration on Subaru) and WFIRST (Wide Field Infra-Red Survey Telescope) are the next generation near-infrared instruments that have a large field-of-view. They allow us to conduct deep and wide transient surveys in near-infrared. Such a near-infrared transient survey enables us to find very distant supernovae that are redshifted to the near-infrared wavelengths. We have performed the mock transient surveys with ULTIMATE-Subaru and WFIRST to investigate their ability to discover Population III pair-instabili…
