We introduce SuperNNova, an open source supernova photometric classification framework which leverages recent advances in deep neural networks. Our core algorithm is a recurrent neural network (RNN) that is trained to classify light-curves using photometric information only. Additional information such as host-galaxy redshift can be incorporated to improve performance. We evaluate our framework using realistic supernovae simulations that include survey detection. We show that our method, for the type Ia vs. non Ia supernovae classification problem, reaches accuracies greater than 96.92 +- 0.09 without any redshift information and up to 99.55 +- 0.06 when redshift, either photometric or spectroscopic, is available. Further, we show that our method attains unprecedented performance for classification of incomplete light-curves, reaching accuracies >86.4 +- 0.1 (>93.5 +- 0.8) without host-galaxy redshift (with redshift information) two days before maximum light. In contrast with previous methods, there is no need for time-consuming feature engineering and we show that our method scales to very large datasets with a modest computing budget. In addition, we investigate often neglected pitfalls of machine learning algorithms. We show that commonly used algorithms suffer from poor calibration and overconfidence on out-of-distribution samples when applied to supernovae data. We devise extensive tests to estimate the robustness of classifiers and cast the learning procedure under a Bayesian light, demonstrating a much better handling of uncertainties. We study the benefits of Bayesian RNNs for SN Ia cosmology. Our code is open-sourced and available on https://github.com/supernnova/SuperNNova.
Amortized Bayesian Inference for Supernovae in the Era of the Vera Rubin Observatory Using Normalizing Flows
Vera Rubin天文台将于2024年年中开始观测,这将使我们的超新星发现率提高到每年超过100万颗。人们一直在大力推动开发新的方法来识别、分类并最终理解鲁宾天文台发现的数百万颗超新星。在这里,我们提出了第一个基于模拟的推理方法,使用归一化流,训练快速推断多变量,Rubin类数据流中的玩具超新星模型的参数。我们发现,我们的方法是很好的校准相比,传统的推理方法(特别是MCMC),在测试时间只需要万分之一的CPU小时。
The Dark Energy Survey 5-year photometrically identified Type Ia Supernovae
在暗能量巡天(DES)中,作为使用Ia型超新星(SN Ia)进行宇宙学分析的一部分,我们使用多波段光变曲线和宿主星系红移来呈现光度学识别的SN Ia样本。为了进行这种分析,我们使用了光度分类框架SuperNNova(SNN; Möller et al. 2019),该框架在逼真的DES模拟上进行了训练。为了可靠的分类,我们处理DES SN程序(DES-SN)数据,并对分类器架构进行改进,在模拟中获得了98%以上的分类精度。这是第一个使用集成方法的SN分类,从而产生更强大的样本。利用测光、主星系红移和分类概率要求,我们识别出1,863个Ia型超新星,并从中选出1,484个宇宙学级的Ia型超新星,其红移范围为0.07 < z < 1.14。我们发现光度选择的样品和模拟的光变曲线特性之间有很好的一致性。此外,我们使用两种类型的贝叶斯神经网络分类器,提供分类概率的不确定性,创建类似的SN Ia样本。我们测试的可行性,使用这些不确定性作为指标的分布的候选人和模型的信心。最后,我们讨论了光度样本和分类方法的影响,为未来的调查,如维拉C。Rubin Observatory Legacy Survey of Space and Time(鲁宾天文台时空遗产巡天)
The Bayesian view of DESI DR2 with unimpeded: Evidence and tension in a combined analysis with CMB and supernovae across cosmological models
Kavli Institute for Cosmology, University of Cambridge(卡弗里宇宙学研究所,剑桥大学);Cavendish Laboratory, University of Cambridge(卡文迪许实验室,剑桥大学);Institute of Astronomy, University of Cambridge(天文学研究所,剑桥大学)
Supernova Light Curves Approximation based on Neural Network Models
HSE University(高等经济大学);Lomonosov Moscow State University(罗蒙诺索夫莫斯科国立大学);Moscow Institute of Physics and Technology(莫斯科物理技术学院);Moscow Polytechnic University(莫斯科理工大学);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)
Finding Strongly Lensed Supernovae from Blended Light Curves
Korea Astronomy and Space Science Institute (KASI)(韩国天体宇宙科学研究所);University of Science and Technology(科学技术联合大学院大学);Lawrence Berkeley National Laboratory(劳伦斯伯克利国家实验室);Berkeley Center for Cosmological Physics & Berkeley Lab, University of California, Berkeley(加州大学伯克利分校伯克利宇宙物理中心与伯克利实验室);Department of Physics & Astronomy, University of San Francisco(旧金山大学物理与天文系)
NMMA-Astro-COLIBRI: An Automated Light-Curve Supernovae Classification Service in the Multi-Survey Era
Université Paris-Saclay – CEA – Irfu(巴黎萨克雷大学);Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Artemis(蔚蓝海岸大学);Laboratoire de Physique et de Chimie de l’Environnement, Université Joseph KI-ZERBO(约瑟夫基泽博大学);Ecole d’ingénieurs aéronautique et spatiale (IPSA)(航空与航天工程师学院);Universität Potsdam, Institut für Physik und Astronomie(波茨坦大学)
Classification of the core-collapse supernova explosion mechanism with learned dictionaries
核心坍缩超新星(CCSN)是引力波的主要来源。对它们典型频率的估计使它们成为目前先进的地面探测器网络的完美目标。成功的探测可能通过波形分析揭示潜在的爆炸机制。这一点已经用超新星模型证据提取器(SMEE; Logue et al.(2012))进行了说明,这是一种基于主成分分析和贝叶斯模型选择的算法。在这里,我们提出了一个补充的方法,SMEE的基础上(监督)字典学习,并表明它是能够重建和分类CCSN信号根据其形态。我们的波形信号是从(a)两个公开可用的目录建立从中微子驱动(Mur)和磁旋转(Dim)CCSN爆炸的数值模拟和(b)从第三个模拟正弦高斯(SG)波形的“模拟”目录。这些信号被注入有色高斯噪声中,以模拟先进LIGO宽带配置中的背景噪声,并缩放到可自由指定的信噪比(SNR)。我们表明,我们的方法正确地分类信号从所有三个字典。特别是,对于SNR=15-20,我们获得了Dim和SG信号的完美匹配以及Mur信号的约85%的真实分类。这些结果与SMEE报告的相同CCSN信号仅注入一个LIGO探测器时的结果相当。我们讨论了我们的方法的主要局限性以及可能的改进。
Dark energy reconstruction analysis with artificial neural networks: Application on simulated Supernova Ia data from Rubin Observatory
National Center for Supercomputing Applications, University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校国家超级计算应用中心);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Kazakh-British Technical University(哈萨克-英国技术大学);Instituto de Ciencias Físicas, Universidad Nacional Autónoma de México(墨西哥国立自治大学物理科学研究所);University of Geneva(日内瓦大学);University of Thessaly(塞萨利大学)
CIGaRS I: Combined simulation-based inference from type Ia supernovae and host photometry
Theoretical and Scientific Data Science, International School for Advanced Studies (SISSA), Trieste, Italy(1 理论与科学数据科学,国际先进研究学院(SISSA),意大利特里este);Institute of Cosmos Sciences (ICC), University of Barcelona, Barcelona, Spain(2 大学巴塞罗那宇宙科学研究所(ICC),西班牙巴塞罗那);Department of Physics, Imperial College London, London, UK(3 物理系,伦敦帝国理工学院,英国伦敦);Institute for Fundamental Physics of the Universe (IFPU), Trieste, Italy(4 宇宙基本物理研究所(IFPU),意大利特里este);Italian Research Center on High-Performance Computing, Big Data and Quantum Computing (ICSC), Casalecchio di Reno, Italy(5 意大利高性能计算、大数据和量子计算研究中心(ICSC),意大利Casalecchio di Reno);Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain(6 哥伦布研究与高级研究机构(ICREA),西班牙巴塞罗那)