Abstract:Type Ia supernovae (SNe Ia) are standarizable candles whose observed light curves can be used to infer their distances, which can in turn be used in cosmological analyses. As the quantity of observed SNe Ia grows with current and upcoming surveys, increasingly scalable analyses are necessary to take full advantage of these new datasets for precise estimation of cosmological parameters. Bayesian inference methods enable fitting SN Ia light curves with robust uncertainty quantification, but traditional posterior sampling using Markov Chain Monte Carlo (MCMC) is computationally expensive. We present an implementation of variational inference (VI) to accelerate the fitting of SN Ia light curves using the BayeSN hierarchical Bayesian model for time-varying SN Ia spectral energy distributions (SEDs). We demonstrate and evaluate its performance on both simulated light curves and data from the Foundation Supernova Survey with two different forms of surrogate posterior -- a multivariate normal and a custom multivariate zero-lower-truncated normal distribution -- and compare them with the Laplace Approximation and full MCMC analysis. To validate of our variational approximation, we calculate the pareto-smoothed importance sampling (PSIS) diagnostic, and perform variational simulation-based calibration (VSBC). The VI approximation achieves similar results to MCMC but with an order-of-magnitude speedup for the inference of the photometric distance moduli. Overall, we show that VI is a promising method for scalable parameter inference that enables analysis of larger datasets for precision cosmology.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to increase the speed of estimating the luminosity distance of Type Ia supernovae (SNe Ia) while maintaining high precision. Specifically, the authors focus on how to use variational inference (VI) to accelerate the time - varying SNe Ia spectral energy distribution (SEDs) fitting process based on the BayeSN hierarchical Bayesian model. Although traditional posterior sampling methods such as Markov chain Monte Carlo (MCMC) can provide robust uncertainty quantification, they are computationally expensive and difficult to handle the increasing amount of data in current and future large - scale astronomical surveys. Therefore, by introducing the VI method, the paper aims to achieve results similar to MCMC but with a ten - fold increase in computational speed, thereby providing a more efficient means of parameter inference for precise cosmological analysis. In the paper, the authors implemented two different forms of approximate posterior distributions - the multivariate normal distribution and a custom - made multivariate zero - lower - bound truncated normal distribution, and evaluated their performance on simulated light curves and Foundation supernova survey data. To verify the effectiveness of the variational approximation, the author also calculated the Pareto - smoothed importance sampling (PSIS) diagnostic index and carried out variational simulation - based calibration (VSBC). The results show that the VI method has achieved results similar to MCMC in terms of inferring the photometric distance modulus, but with a ten - fold increase in speed, demonstrating that VI is a promising method for large - scale data analysis and is suitable for processing large data sets for precise cosmological research.

Variational Inference for Acceleration of SN Ia Photometric Distance Estimation with BayeSN

Variational Inference for Acceleration of SN Ia Photometric Distance Estimation with BayeSN

Scalable hierarchical BayeSN inference: investigating dependence of SN Ia host galaxy dust properties on stellar mass and redshift

A BayeSN Distance Ladder: $H_0$ from a consistent modelling of Type Ia supernovae from the optical to the near infrared

SIDE-real: Supernova Ia Dust Extinction with truncated marginal neural ratio estimation applied to real data

Accounting for Selection Effects in Supernova Cosmology with Simulation-Based Inference and Hierarchical Bayesian Modelling

SICRET: Supernova Ia Cosmology with truncated marginal neural Ratio EsTimation

Faster Bayesian inference with neural network bundles and new results for models

Bird-Snack: Bayesian Inference of dust law $R_V$ Distributions using SN Ia Apparent Colours at peaK

Measuring the Ejecta Velocities of Type Ia Supernovae from the Pan-STARRS1 Medium Deep Survey

Union Through UNITY: Cosmology with 2,000 SNe Using a Unified Bayesian Framework

STAR NRE: Solving supernova selection effects with set-based truncated auto-regressive neural ratio estimation

Faster Bayesian inference with neural network bundles and new results for $f(R)$ models

Extrapolation of Type Ia Supernova Spectra into the Near-Infrared Using PCA

Unsupervised Transient Light Curve Analysis Via Hierarchical Bayesian Inference

Connecting Infrared Surface Brightness Fluctuation Distances to Type Ia Supernova Hosts: Testing the Top Rung of the Distance Ladder

Kilonova Light Curve Parameter Estimation Using Likelihood-Free Inference

Photo-zSNthesis: Converting Type Ia Supernova Lightcurves to Redshift Estimates via Deep Learning