Development of an Ultra-fast, Likelihood-based, Distance Inference Framework for the Next Generation of Type Ia Supernova Surveys

Abstract

In this work, we present EDRIS (French for Distance Estimator for Incomplete Supernova Surveys), a cosmological inference framework tailored to reconstruct unbiased cosmological distances from type Ia supernovae light-curve parameters. This goal is achieved by including data truncation directly in the statistical model which takes care of the standardization of luminosity distances. It allows us to build a single-step distance estimate by maximizing the corresponding likelihood, free from the biases the survey detection limits would introduce otherwise. Moreover, we expect the current worldwide statistics to be multiplied by O(10) in the upcoming years. This provides a new challenge to handle as the cosmological analysis must stay computationally towable. We show that the optimization methods used in EDRIS allow for a reasonable time complexity of O($N^2$) resulting in a very fast inference process (O(10s) for 1500 supernovae).