Unsupervised machine learning for transient discovery in Deeper, Wider, Faster light curves

Abstract

Identification of anomalous light curves within time-domain surveys is often challenging. In addition, with the growing number of wide-field surveys and the volume of data produced exceeding astronomers ability for manual evaluation, outlier and anomaly detection is becoming vital for transient science. We present an unsupervised method for transient discovery using a clustering technique and the Astronomaly package. As proof of concept, we evaluate 85553 minute-cadenced light curves collected over two 1.5 hour periods as part of the Deeper, Wider, Faster program, using two different telescope dithering strategies. By combining the clustering technique HDBSCAN with the isolation forest anomaly detection algorithm via the visual interface of Astronomaly, we are able to rapidly isolate anomalous sources for further analysis. We successfully recover the known variable sources, across a range of catalogues from within the fields, and find a further 7 uncatalogued variables and two stellar flare events, including a rarely observed ultra fast flare (5 minute) from a likely M-dwarf.

Accepted 7 Aug 2020, 19 pages, 8 figures,

References in corpus (53)

• Scikit-learn: Machine Learning in Python
• The Astropy Project: Building an inclusive, open-science project and status of the v2.0 core package
• Measuring Reddening with SDSS Stellar Spectra and Recalibrating SFD
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• The Zwicky Transient Facility: System Overview, Performance, and First Results
• A bright millisecond radio burst of extragalactic origin
• The Man Behind the Curtain: X-rays Drive the UV through NIR Variability in the 2013 AGN Outburst in NGC 2617
• The K2 Mission: Characterization and Early results
• The Dark Energy Camera
• The Dark Energy Survey: more than dark energy - an overview
• LSST Science Book, Version 2.0
• The Zwicky Transient Facility: Science Objectives
• Gaia Data Release 2: Photometric content and validation
• Gaia Data Release 2: using Gaia parallaxes
• The Pan-STARRS1 Surveys
• The TESS Input Catalog and Candidate Target List
• On Machine-Learned Classification of Variable Stars with Sparse and Noisy Time-Series Data
• SkyMapper Southern Survey: Second Data Release (DR2)
• The Cow: discovery of a luminous, hot and rapidly evolving transient
• The Fast, Luminous Ultraviolet Transient AT2018cow: Extreme Supernova, or Disruption of a Star by an Intermediate-Mass Black Hole?
• Follow up of GW170817 and its electromagnetic counterpart by Australian-led observing programs
• Automated supervised classification of variable stars I. Methodology
• Photometric Supernova Classification With Machine Learning
• SuperNNova: an open-source framework for Bayesian, Neural Network based supernova classification
• Automating Discovery and Classification of Transients and Variable Stars in the Synoptic Survey Era
• General relativistic orbital decay in a seven-minute-orbital-period eclipsing binary system
• RAPID: Early Classification of Explosive Transients using Deep Learning
• Gaia Data Release 2: Variable stars in the colour-absolute magnitude diagram
• Precise Throughput Determination of the PanSTARRS Telescope and the Gigapixel Imager using a Calibrated Silicon Photodiode and a Tunable Laser: Initial Results
• QSO Selection Algorithm Using Time Variability and Machine Learning: Selection of 1,620 QSO Candidates from MACHO LMC Database
• Machine Learning-based Brokers for Real-time Classification of the LSST Alert Stream
• The EPOCH Project: I. Periodic variable stars in the EROS-2 LMC database
• Shock Breakout and Early Light Curves of Type II-P Supernovae Observed with Kepler
• A Package for the Automated Classification of Periodic Variable Stars
• 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
• Deep-Learnt Classification of Light Curves
• A Search for Fast Optical Transients in the Pan-STARRS1 Medium-Deep Survey: M Dwarf Flares, Asteroids, Limits on Extragalactic Rates, and Implications for LSST
• Systematic Serendipity: A Test of Unsupervised Machine Learning as a Method for Anomaly Detection
• Short Duration Stellar Flares in GALEX Data
• Automatic Classification of Variable Stars in Catalogs with missing data
• Probing the extragalactic fast transient sky at minute timescales with DECam
• An improved quasar detection method in EROS-2 and MACHO LMC datasets
• Clustering Based Feature Learning on Variable Stars
• Unsupervised Classification of Variable Stars
• Optical Observations of Gamma-Ray Bursts, the Discovery of Supernovae 2005bv, 2005ee, and 2006ak, and Searches for Transients Using the "MASTER" Robotic Telescope
• FATS: Feature Analysis for Time Series
• Identification of RR Lyrae stars in multiband, sparsely-sampled data from the Dark Energy Survey using template fitting and Random Forest classification
• Mary, a pipeline to aid discovery of optical transients
• An optical search for transients lasting a few seconds
• Enabling near real-time remote search for fast transient events with lossy data compression
• A Novel, Fully Automated Pipeline for Period Estimation in the EROS 2 Data Set
• Collaborative workspaces to accelerate discovery

Cited by in corpus (18)

• Astronomaly: Personalised Active Anomaly Detection in Astronomical Data
• Anomaly detection in the Zwicky Transient Facility DR3
• A review of unsupervised learning in astronomy
• A unique, ring-like radio source with quadrilateral structure detected with machine learning
• Real-Time Detection of Anomalies in Large-Scale Transient Surveys
• Deep Attention-Based Supernovae Classification of Multi-Band Light-Curves
• Rapid sorting of radio galaxy morphology using Haralick features
• Enabling Unsupervised Discovery in Astronomical Images through Self-Supervised Representations
• The Deeper, Wider, Faster Program: Exploring stellar flare activity with deep, fast cadenced DECam imaging via machine learning
• Identifying outliers in astronomical images with unsupervised machine learning
• Semi-Supervised Classification and Clustering Analysis for Variable Stars
• A Classifier-Based Approach to Multi-Class Anomaly Detection for Astronomical Transients
• Automated classification of eclipsing binary systems in the VVV Survey
• Finding radio transients with anomaly detection and active learning based on volunteer classifications
• The ASKAP Variables and Slow Transients (VAST) Pilot Survey
• Learning to Detect Interesting Anomalies
• Finding Fast Transients in Real Time Using Novel Light Curve Analysis Algorithm
• Enabling New Discoveries with Machine Learning