Real-Time Active Learning for optimised spectroscopic follow-up: Enhancing early SN Ia classification with the Fink broker

Abstract

Current and future surveys rely on machine learning classification to obtain large and complete samples of transients. Many of these algorithms are restricted by training samples that contain a limited number of spectroscopically confirmed events. Here, we present the first real-time application of Active Learning to optimise spectroscopic follow-up with the goal of improving training sets of early type Ia supernovae (SNe Ia) classifiers. Using a photometric classifier for early SN Ia, we apply an Active Learning strategy for follow-up optimisation using the real-time FINK broker processing of the ZTF public stream. We perform follow-up observations at the ANU 2.3m telescope in Australia and obtain 92 spectroscopic classified events that are incorporated in our training set. We show that our follow-up strategy yields a training set that, with 25% less spectra, improves classification metrics when compared to publicly reported spectra. Our strategy selects in average fainter events and, not only supernovae types, but also microlensing events and flaring stars which are usually not incorporated on training sets. Our results confirm the effectiveness of active learning strategies to construct optimal training samples for astronomical classifiers. With the Rubin Observatory LSST soon online, we propose improvements to obtain earlier candidates and optimise follow-up. This work paves the way to the deployment of real-time AL follow-up strategies in the era of large surveys.

Accepted for publication in PASA. 14 pages, 6 figures

References in corpus (29)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• The Zwicky Transient Facility: System Overview, Performance, and First Results
• Overview of the DESI Legacy Imaging Surveys
• SALT2: using distant supernovae to improve the use of Type Ia supernovae as distance indicators
• Rates and properties of type Ia supernovae as a function of mass and star-formation in their host galaxies
• SNANA: A Public Software Package for Supernova Analysis
• The Wide Field Spectrograph (WiFeS)
• Gemini Spectroscopy of Supernovae from SNLS: Improving High Redshift SN Selection and Classification
• The Automatic Learning for the Rapid Classification of Events (ALeRCE) Alert Broker
• SuperNNova: an open-source framework for Bayesian, Neural Network based supernova classification
• Supernova Simulations and Strategies For the Dark Energy Survey
• Results from the Supernova Photometric Classification Challenge
• SALT3: An Improved Type Ia Supernova Model for Measuring Cosmic Distances
• MARZ: Manual and Automatic Redshifting Software
• Fink, a new generation of broker for the LSST community
• Transient processing and analysis using $\text{AMPEL}$: Alert Management, Photometry and Evaluation of Lightcurves
• Astronomaly: Personalised Active Anomaly Detection in Astronomical Data
• Optimizing spectroscopic follow-up strategies for supernova photometric classification with active learning
• Kernel PCA for type Ia supernovae photometric classification
• Spectrophotometric templates for core collapse supernovae and their application in simulations of time-domain surveys
• The ANTARES Astronomical Time-Domain Event Broker
• Photometric classification of type Ia supernovae in the SuperNova Legacy Survey with supervised learning
• Searching for changing-state AGNs in massive datasets -- I: applying deep learning and anomaly detection techniques to find AGNs with anomalous variability behaviours
• Active Anomaly Detection for time-domain discoveries
• The Dark Energy Survey 5-year photometrically identified Type Ia Supernovae
• Fink: early supernovae Ia classification using active learning
• ATAT: Astronomical Transformer for time series And Tabular data
• Enabling the discovery of fast transients: A kilonova science module for the Fink broker
• Optimising a magnitude-limited spectroscopic training sample for photometric classification of supernovae

Cited by in corpus (2)

• Early Identification of Optical Tidal Disruption Events: A science module for the Fink broker
• LStein: A new approach to visualizing sparse 2.5-dimensional data