RAPID: Early Classification of Explosive Transients using Deep Learning

Abstract

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 the data, making RAPID well-suited for processing the millions of alerts that ongoing and upcoming wide-field surveys such as the Zwicky Transient Facility (ZTF), and the Large Synoptic Survey Telescope (LSST) will produce. The classification accuracy improves over the lifetime of the transient as more photometric data becomes available, and across the 12 transient classes, we obtain an average area under the receiver operating characteristic curve of 0.95 and 0.98 at early and late epochs, respectively. We demonstrate RAPID's ability to effectively provide early classifications of observed transients from the ZTF data stream. We have made RAPID available as an open-source software package (https://astrorapid.readthedocs.io) for machine learning-based alert-brokers to use for the autonomous and quick classification of several thousand light curves within a few seconds.

Accepted version. 28 pages, 16 figures, 2 tables, PASP Special Issue on Methods for Time-Domain Astrophysics. Submitted: 13 December 2018, Accepted: 26 March 2019

References in corpus (63)

• Astropy: A Community Python Package for Astronomy
• emcee: The MCMC Hammer
• The NumPy array: a structure for efficient numerical computation
• GW170817: Observation of Gravitational Waves from a Binary Neutron Star Inspiral
• Multi-messenger Observations of a Binary Neutron Star Merger
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• Correcting for the Effects of Interstellar Extinction
• ATLAS: A High-Cadence All-Sky Survey System
• Origin of the heavy elements in binary neutron-star mergers from a gravitational wave event
• Swope Supernova Survey 2017a (SSS17a), the Optical Counterpart to a Gravitational Wave Source
• The Dark Energy Survey: more than dark energy - an overview
• SALT2: using distant supernovae to improve the use of Type Ia supernovae as distance indicators
• Supernova Light Curves Powered by Young Magnetars
• Exploring the Optical Transient Sky with the Palomar Transient Factory
• Optical emission from a kilonova following a gravitational-wave-detected neutron-star merger
• The Electromagnetic Counterpart of the Binary Neutron Star Merger LIGO/Virgo GW170817. I. Dark Energy Camera Discovery of the Optical Counterpart
• The Supernova Legacy Survey 3-year sample: Type Ia Supernovae photometric distances and cosmological constraints
• Real-Time Difference Imaging Analysis of MOA Galactic Bulge Observations During 2000
• Kilonovae
• Faint Thermonuclear Supernovae from AM Canum Venaticorum Binaries
• Type Iax Supernovae: A New Class of Stellar Explosion
• Berkeley Supernova Ia Program I: Observations, Data Reduction, and Spectroscopic Sample of 582 Low-Redshift Type Ia Supernovae
• SNANA: A Public Software Package for Supernova Analysis
• The magnetar model for Type I superluminous supernovae I: Bayesian analysis of the full multicolour light curve sample with MOSFiT
• Expanding the Realm of Microlensing Surveys with Difference Image Photometry
• SN 2005ap: A Most Brilliant Explosion
• Pair Instability Supernovae: Light Curves, Spectra, and Shock Breakout
• Weighing Black Holes using Tidal Disruption Events
• The Sloan Digital Sky Survey-II Supernova Survey: Search Algorithm and Follow-up Observations
• MOSFiT: Modular Open-Source Fitter for Transients
• Calcium-rich gap transients in the remote outskirts of galaxies
• Photometric Supernova Classification With Machine Learning
• Thermonuclear .Ia Supernovae from Helium Shell Detonations: Explosion Models and Observables
• Results from the Supernova Photometric Classification Challenge
• Automating Discovery and Classification of Transients and Variable Stars in the Synoptic Survey Era
• An Intermediate Luminosity Transient in NGC300: The Eruption of a Dust-Enshrouded Massive Star
• Photometric SN Ia Candidates from the Three-Year SDSS-II SN Survey Data
• A recurrent neural network for classification of unevenly sampled variable stars
• A long-lived remnant neutron star after GW170817 inferred from its associated kilonova
• Deep Recurrent Neural Networks for Supernovae Classification
• Photometric selection of high-redshift type Ia supernovae
• Machine Learning-based Brokers for Real-time Classification of the LSST Alert Stream
• Analysis of RR Lyrae Stars in the Northern Sky Variability Survey
• Testing Models of Intrinsic Brightness Variations in Type Ia Supernovae, and their Impact on Measuring Cosmological Parameters
• Theoretical Models of Optical Transients. I. A Broad Exploration of the Duration-Luminosity Phase Space
• Two New Calcium-Rich Gap Transients in Group and Cluster Environments
• Optimizing spectroscopic follow-up strategies for supernova photometric classification with active learning
• Kernel PCA for type Ia supernovae photometric classification
• A simple and robust method for automated photometric classification of supernovae using neural networks
• Photometric classification of type Ia supernovae in the SuperNova Legacy Survey with supervised learning
• Bayesian Single-Epoch Photometric Classification of Supernovae
• Machine Learning Techniques for Stellar Light Curve Classification
• PELICAN: deeP architecturE for the LIght Curve ANalysis
• Semi-supervised Learning for Photometric Supernova Classification
• Calcium-Rich Gap Transients: Tidal Detonations of White Dwarfs?
• Deep multi-survey classification of variable stars
• STACCATO: A Novel Solution to Supernova Photometric Classification with Biased Training Sets
• Classifying Supernovae Using Only Galaxy Data
• SIDRA: a blind algorithm for signal detection in photometric surveys
• Statistical Classification Techniques for Photometric Supernova Typing
• ANTARES: A Prototype Transient Broker System
• ANTARES: Progress towards building a `Broker' of time-domain alerts
• An intermediate-luminosity-optical-transient (ILOT) model for the young stellar object ASASSN-15qi

Cited by in corpus (94)

• Design and operation of the ATLAS Transient Science Server
• The Zwicky Transient Facility Bright Transient Survey. II. A Public Statistical Sample for Exploring Supernova Demographics
• The Automatic Learning for the Rapid Classification of Events (ALeRCE) Alert Broker
• Alert Classification for the ALeRCE Broker System: The Light Curve Classifier
• GROWTH on S190425z: Searching thousands of square degrees to identify an optical or infrared counterpart to a binary neutron star merger with the Zwicky Transient Facility and Palomar Gattini IR
• Fink, a new generation of broker for the LSST community
• The Young Supernova Experiment: Survey Goals, Overview, and Operations
• GRANDMA Observations of Advanced LIGO's and Advanced Virgo's Third Observational Campaign
• SuperRAENN: A Semi-supervised Supernova Photometric Classification Pipeline Trained on Pan-STARRS1 Medium Deep Survey Supernovae
• Fast-transient Searches in Real Time with ZTFReST: Identification of Three Optically-discovered Gamma-ray Burst Afterglows and New Constraints on the Kilonova Rate
• Supernova Host Galaxies in the Dark Energy Survey: I. Deep Coadds, Photometry, and Stellar Masses
• DASH: Deep Learning for the Automated Spectral Classification of Supernovae and their Hosts
• Double-, triple-line spectroscopic candidates in the LAMOST Medium-Resolution Spectroscopic Survey
• A Deep Learning Approach for Active Anomaly Detection of Extragalactic Transients
• The Young Supernova Experiment Data Release 1 (YSE DR1): Light Curves and Photometric Classification of 1975 Supernovae
• Superluminous supernovae: an explosive decade
• GHOST: Using Only Host Galaxy Information to Accurately Associate and Distinguish Supernovae
• On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
• The Dark Energy Survey 5-year photometrically identified Type Ia Supernovae
• ParSNIP: Generative Models of Transient Light Curves with Physics-Enabled Deep Learning
• FLEET: A Redshift-Agnostic Machine Learning Pipeline to Rapidly Identify Hydrogen-Poor Superluminous Supernovae
• Unsupervised machine learning for transient discovery in Deeper, Wider, Faster light curves
• Efficiently jet-powered radiation in intermediate-luminosity optical transients (ILOTS)
• Machine Learning for Observational Cosmology
• The ZTF Source Classification Project: I. Methods and Infrastructure
• Identifying Tidal Disruption Events with an Expansion of the FLEET Machine Learning Algorithm
• Light curve classification with recurrent neural networks for GOTO: dealing with imbalanced data
• A Method to Distinguish Quiescent and Dusty Star-forming Galaxies with Machine Learning
• The first Hubble diagram and cosmological constraints using superluminous supernova
• Deep Attention-Based Supernovae Classification of Multi-Band Light-Curves
• Optimizing Serendipitous Detections of Kilonovae: Cadence and Filter Selection
• Background Rejection in Atmospheric Cherenkov Telescopes using Recurrent Convolutional Neural Networks
• Optimizing Cadences with Realistic Light Curve Filtering for Serendipitous Kilonova Discovery with Vera Rubin Observatory
• Photometric Classification of 2315 Pan-STARRS1 Supernovae with Superphot
• Vetting the optical transient candidates detected by the GWAC network using convolutional neural networks
• Real-Time Detection of Anomalies in Large-Scale Transient Surveys
• Spectroscopic Studies of Type Ia Supernovae Using LSTM Neural Networks
• SNIascore: Deep Learning Classification of Low-Resolution Supernova Spectra
• Photometric Classification of Early-Time Supernova Lightcurves with SCONE
• Real-Time Value-Driven Data Augmentation in the Era of LSST
• Measuring the Hubble Constant with cosmic chronometers: a machine learning approach
• MeerCRAB: MeerLICHT Classification of Real and Bogus Transients using Deep Learning
• LSST Cadence Strategy Evaluations for AGN Time-series Data in Wide-Fast-Deep Field
• ATAT: Astronomical Transformer for time series And Tabular data
• Identify Light-Curve Signals with Deep Learning Based Object Detection Algorithm. I. Transit Detection
• Pan-chromatic photometric classification of supernovae from multiple surveys and transfer learning for future surveys
• Explaining recently studied intermediate luminosity optical transients (ILOTs) with jet powering
• deepSIP: Linking Type Ia Supernova Spectra to Photometric Quantities with Deep Learning
• A classification algorithm for time-domain novelties in preparation for LSST alerts: Application to variable stars and transients detected with DECam in the Galactic Bulge
• Star formation and morphological properties of galaxies in the Pan-STARRS $3\pi$ survey- I. A machine learning approach to galaxy and supernova classification
• $\text{tdescore}$: An Accurate Photometric Classifier for Tidal Disruption Events
• Intermediate Luminosity Optical Transients (ILOTs) from merging giants
• Classifying Astronomical Transients Using Only Host Galaxy Photometry
• Enabling the discovery of fast transients: A kilonova science module for the Fink broker
• TiDES: The 4MOST Time Domain Extragalactic Survey
• HAFFET: Hybrid Analytic Flux FittEr for Transients
• Photometric classification of HSC transients using machine learning
• Identifying Light-curve Signals with a Deep Learning Based Object Detection Algorithm. II. A General Light Curve Classification Framework
• Deep learning method in testing the cosmic distance duality relation
• Optimising a magnitude-limited spectroscopic training sample for photometric classification of supernovae
• SNAD Transient Miner: Finding Missed Transient Events in ZTF DR4 using k-D trees
• Data-driven detection of multi-messenger transients
• Impact of Rubin Observatory cadence choices on supernovae photometric classification
• The circumstellar matter of type II intermediate luminosity optical transients (ILOTs)
• A Classifier-Based Approach to Multi-Class Anomaly Detection for Astronomical Transients
• Results from the Pan-STARRS Search for Kilonovae: Contamination by Massive Stellar Outbursts
• SNGuess: A method for the selection of young extragalactic transients
• Photometric Data-driven Classification of Type Ia Supernovae in the Open Supernova Catalog
• On the nature of the planet-powered transient event ZTF SLRN-2020
• ORACLE: A Real-Time, Hierarchical, Deep-Learning Photometric Classifier for the LSST
• Transfer Learning for Transient Classification: From Simulations to Real Data and ZTF to LSST
• A catalog to unite them all: REGALADE, a revised galaxy compilation for the advanced detector era
• Forecasting Supernova Observations with the CSST: I. Photometric Samples
• Searching for Quasi-Periodic Eruptions using Machine Learning
• The ATLAS Virtual Research Assistant
• A rapidly fading star as a type II obscuring intermediate luminosity optical transient (ILOT) in a triple star system
• On the nature of jets from a main sequence companion at the onset of common envelope evolution
• STag: Supernova Tagging and Classification
• Testing and Combining Transient Spectral Classification Tools on 4MOST-like Blended Spectra
• Leveraging pre-trained vision Transformers for multi-band photometric light curve classification
• FALCO: a Foundation model of Astronomical Light Curves for time dOmain astronomy
• Kilonova-Targeting Lightcurve Classification for Wide Field Survey Telescope
• Building spectral templates and reconstructing parameters for core collapse supernovae with CASTOR
• Real-time Light Curve Classification Framework for the Wide Field Survey Telescope Using Modified Semi-supervised Variational Auto-Encoder
• The automation of optical transient discovery and classification in Rubin-era time-domain astronomy
• Harvesting the Lyα forest with convolutional neural networks
• El-CID: A filter for Gravitational-wave Electromagnetic Counterpart Identification
• The snapshot distance method: estimating the distance to a Type Ia supernova from minimal observations
• Considerations for optimizing photometric classification of supernovae from the Rubin Observatory
• AMPEL workflows for LSST: Modular and reproducible real-time photometric classification
• A Template-Based Approach to the Photometric Classification of SN 1991bg-like Supernovae in the SDSS-II Supernova Survey
• Modeling Irregular Astronomical Time Series with Neural Stochastic Delay Differential Equations
• Faint intermediate luminosity optical transients (ILOTs) from engulfing exoplanets on the Hertzsprung gap
• PIE-ADA: Physics-Informed Ensemble with Adaptive Data Augmentation for Photometric Transient Classification