ParSNIP: Generative Models of Transient Light Curves with Physics-Enabled Deep Learning

Abstract

We present a novel method to produce empirical generative models of all kinds of astronomical transients from datasets of unlabeled light curves. Our hybrid model, that we call ParSNIP, uses a neural network to model the unknown intrinsic diversity of different transients and an explicit physics-based model of how light from the transient propagates through the universe and is observed. The ParSNIP model predicts the time-varying spectra of transients despite only being trained on photometric observations. With a three-dimensional intrinsic model, we are able to fit out-of-sample multiband light curves of many different kinds of transients with model uncertainties of 0.04-0.06 mag. The representation learned by the ParSNIP model is invariant to redshift, so it can be used to perform photometric classification of transients even with heavily biased training sets. Our classification techniques significantly outperform state-of-the-art methods on both simulated (PLAsTiCC) and real (PS1) datasets with 2.3$\times$ and 2$\times$ less contamination respectively for classification of Type~Ia supernovae. We demonstrate how our model can identify previously-unobserved kinds of transients and produce a sample that is 90% pure. The ParSNIP model can also estimate distances to Type Ia supernovae in the PS1 dataset with an RMS of 0.150 $\pm$ 0.007 mag compared to 0.155 $\pm$ 0.008 mag for the SALT2 model on the same sample. We discuss how our model could be used to produce distance estimates for supernova cosmology without the need for explicit classification.

27 pages, 17 figures, accepted for publication in AJ

References in corpus (14)

• Array Programming with NumPy
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• An Analysis of the Shapes of Interstellar Extinction Curves. V. The IR-Through-UV Curve Morphology
• PESSTO : survey description and products from the first data release by the Public ESO Spectroscopic Survey of Transient Objects
• Recent Advances in Autoencoder-Based Representation Learning
• A recurrent neural network for classification of unevenly sampled variable stars
• PELICAN: deeP architecturE for the LIght Curve ANalysis
• A Deep Learning Approach for Active Anomaly Detection of Extragalactic Transients
• 4MOST Consortium Survey 10: The Time-Domain Extragalactic Survey (TiDES)
• The Twins Embedding of Type Ia Supernovae II: Improving Cosmological Distance Estimates
• The Twins Embedding of Type Ia Supernovae I: The Diversity of Spectra at Maximum Light
• Peculiar velocity cosmology with type Ia supernovae
• Unsupervised Transient Light Curve Analysis Via Hierarchical Bayesian Inference
• Considerations for optimizing photometric classification of supernovae from the Rubin Observatory

Cited by in corpus (7)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• The Young Supernova Experiment Data Release 1 (YSE DR1): Light Curves and Photometric Classification of 1975 Supernovae
• Photometric Classification of Early-Time Supernova Lightcurves with SCONE
• Carnegie Supernova Project-II: Near-infrared spectral diversity and template of Type Ia Supernovae
• Impact of Rubin Observatory cadence choices on supernovae photometric classification
• Deep Learning Unresolved Lensed Lightcurves
• A machine learning approach to assessing the presence of substructure in quasar host galaxies using the Hyper Suprime-Cam Subaru Strategic Program