标题： 下一代Ia型超新星巡天的超快速似然距离推理框架的发展 显示英文标题

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

摘要： 在这项工作中，我们提出了EDRIS（法语中“用于不完整超新星调查的距离估计器”的意思），这是一个宇宙学推理框架，旨在从Ia型超新星光曲线参数重构无偏的宇宙距离。通过直接在统计模型中包括数据截断，该模型负责标准化光度距离，这一目标得以实现。这使我们能够通过最大化相应的似然函数来构建一个单步距离估计，从而避免了调查检测限制否则会引入的偏差。此外，我们预计未来几年全球统计数据将增加O(10)倍。这对宇宙学分析提出了新的挑战，必须保持计算上可行。我们展示了EDRIS中使用的优化方法允许时间复杂度为O($N^2$)，从而实现了非常快速的推理过程（对于1500个超新星大约需要10秒）。 显示英文摘要

摘要： 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).