On the cosmological performance of photometrically classified supernovae with machine learning

Abstract

The efficient classification of different types of supernova is one of the most important problems for observational cosmology. However, spectroscopic confirmation of most objects in upcoming photometric surveys, such as the The Rubin Observatory Legacy Survey of Space and Time (LSST), will be unfeasible. The development of automated classification processes based on photometry has thus become crucial. In this paper we investigate the performance of machine learning (ML) classification on the final cosmological constraints using simulated lightcurves from The Supernova Photometric Classification Challenge, released in 2010. We study the use of different feature sets for the lightcurves and many different ML pipelines based on either decision tree ensembles or automated search processes. To construct the final catalogs we propose a threshold selection method, by employing a \emph{Bias-Variance tradeoff}. This is a very robust and efficient way to minimize the Mean Squared Error. With this method we were able to get very strong cosmological constraints, which allowed us to keep $\sim 75\%$ of the total information in the type Ia SNe when using the SALT2 feature set and $\sim 33\%$ for the other cases (based on either the Newling model or on standard wavelet decomposition).

arXiv comment18 pages, 13 figures, 7 tables. Replaced to match the published version

References in corpus (54)

• Scikit-learn: Machine Learning in Python
• The Complete Light-curve Sample of Spectroscopically Confirmed Type Ia Supernovae from Pan-STARRS1 and Cosmological Constraints from The Combined Pantheon Sample
• A 2.4% Determination of the Local Value of the Hubble Constant
• Improved cosmological constraints from a joint analysis of the SDSS-II and SNLS supernova samples
• The Dark Energy Survey: more than dark energy - an overview
• LSST Science Book, Version 2.0
• SALT2: using distant supernovae to improve the use of Type Ia supernovae as distance indicators
• Improved Distances to Type Ia Supernovae with Multicolor Light Curve Shapes: MLCS2k2
• First-year Sloan Digital Sky Survey-II (SDSS-II) Supernova Results: Hubble Diagram and Cosmological Parameters
• H0LiCOW V. New COSMOGRAIL time delays of HE0435-1223: $H_0$ to 3.8% precision from strong lensing in a flat $\Lambda$CDM model
• Gravitationally lensed quasars and supernovae in future wide-field optical imaging surveys
• H0LiCOW - IX. Cosmographic analysis of the doubly imaged quasar SDSS 1206+4332 and a new measurement of the Hubble constant
• A Critical Assessment of Photometric Redshift Methods: A CANDELS Investigation
• Correcting Type Ia Supernova Distances for Selection Biases and Contamination in Photometrically Identified Samples
• J-PLUS: The Javalambre Photometric Local Universe Survey
• The Sloan Digital Sky Survey-II Supernova Survey: Search Algorithm and Follow-up Observations
• The Data Release of the Sloan Digital Sky Survey-II Supernova Survey
• The Southern Photometric Local Universe Survey (S-PLUS): improved SEDs, morphologies and redshifts with 12 optical filters
• ANNz2 - photometric redshift and probability distribution function estimation using machine learning
• Photometric Supernova Classification With Machine Learning
• The Carnegie Supernova Project: Absolute Calibration and the Hubble Constant
• Probing the anisotropic local universe and beyond with SNe Ia data
• SuperNNova: an open-source framework for Bayesian, Neural Network based supernova classification
• Results from the Supernova Photometric Classification Challenge
• J-PAS: The Javalambre-Physics of the Accelerated Universe Astrophysical Survey
• Measuring Dark Energy Properties with Photometrically Classified Pan-STARRS Supernovae. II. Cosmological Parameters
• The Core-collapse rate from the Supernova Legacy Survey
• A Measurement of the Rate of type-Ia Supernovae at Redshift $z\approx$ 0.1 from the First Season of the SDSS-II Supernova Survey
• A Comparison of Six Photometric Redshift Methods Applied to 1.5 Million Luminous Red Galaxies
• Deep Recurrent Neural Networks for Supernovae Classification
• First Cosmology Results using Type Ia Supernova from the Dark Energy Survey: Simulations to Correct Supernova Distance Biases
• Cosmological Constraints from Type Ia Supernovae Peculiar Velocity Measurements
• Bayesian Estimation Applied to Multiple Species: Towards cosmology with a million supernovae
• Kernel PCA for type Ia supernovae photometric classification
• Supernova Photometric Classification Pipelines Trained on Spectroscopically Classified Supernovae from the Pan-STARRS1 Medium-Deep Survey
• A simple and robust method for automated photometric classification of supernovae using neural networks
• Strongly lensed SNe Ia in the era of LSST: observing cadence for lens discoveries and time-delay measurements
• Measuring the growth rate of structure with Type IA Supernovae from LSST
• On the Possibility of Anisotropic Curvature in Cosmology
• Improving Photometric Redshift Estimation using GPz: size information, post processing and improved photometry
• Accurate Weak Lensing of Standard Candles. II. Measuring sigma8 with Supernovae
• Classifying Supernovae Using Only Galaxy Data
• Percent-Level Test of Isotropic Expansion Using Type Ia Supernovae
• A new sample of southern radio galaxies: Host galaxy masses and star-formation rates
• First measurement of ${\sigma }_8$ using supernova magnitudes only
• Statistical Classification Techniques for Photometric Supernova Typing
• The Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC): Selection of a performance metric for classification probabilities balancing diverse science goals
• Turning noise into signal: learning from the scatter in the Hubble diagram
• Measuring the Hubble function with standard candle clustering
• The effects of velocities and lensing on moments of the Hubble diagram
• Improved Photometric Classification of Supernovae using Deep Learning
• On the amount of peculiar velocity field information in supernovae from LSST and beyond
• Parameter Estimation with BEAMS in the presence of biases and correlations
• Using Random Forest Machine Learning Algorithms in Binary Supernovae Classification

Cited by in corpus (4)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• The miniJPAS survey: star-galaxy classification using machine learning
• J-PLUS DR3: Galaxy-Star-Quasar classification
• Unveiling the Universe with Emerging Cosmological Probes