Blending of galaxies has a major contribution in the systematic error budget of weak lensing studies, affecting photometric and shape measurements, particularly for ground-based, deep, photometric galaxy surveys, such as the Rubin Observatory Legacy Survey of Space and Time (LSST). Existing deblenders mostly rely on analytic modelling of galaxy profiles and suffer from the lack of flexible yet accurate models. We propose to use generative models based on deep neural networks, namely variational autoencoders (VAE), to learn probabilistic models directly from data. We train a VAE on images of centred, isolated galaxies, which we reuse, as a prior, in a second VAE-like neural network in charge of deblending galaxies. We train our networks on simulated images including six LSST bandpass filters and the visible and near-infrared bands of the Euclid satellite, as our method naturally generalises to multiple bands and can incorporate data from multiple instruments. We obtain median reconstruction errors on ellipticities and $r$-band magnitude between $\pm{0.01}$ and $\pm{0.05}$ respectively in most cases, and ellipticity multiplicative bias of 1.6% for blended objects in the optimal configuration. We also study the impact of decentring and prove the method to be robust. This method only requires the approximate centre of each target galaxy, but no assumptions about the number of surrounding objects, pointing to an iterative detection/deblending procedure we leave for future work. Finally, we discuss future challenges about training on real data and obtain encouraging results when applying transfer learning. Our code is publicly available on GitHub (https://github.com/LSSTDESC/DeblenderVAE).
Multiband Probabilistic Cataloging: A Joint Fitting Approach to Point Source Detection and Deblending
概率编目（PCAT）在拥挤的领域中优于传统的单波段光学数据编目方法（Portillo et al. 2017）。我们将我们的工作扩展到多个波段，与以前的单波段结果相比，实现了更高的灵敏度（$\sim$ 0.4 mag）和更高的速度（500倍）。我们证明了多波段PCAT模拟数据的有效性，无论是在恢复准确的后验在目录空间，并在直接去混源。以Sloan数字巡天（SDSS）观测的M2为例，以Hubble空间望远镜数据为真值，对$r$和$i$波段数据的联合拟合比单波段概率星表法更深入$\sim0.4$，对F606 W $\leq 20$的假发现率小于20%.与DAOPHOT相比，使用相同的数据，双波段SDSS星表拟合深度接近1.5个星等，并保持了较低的错误发现率，直到F606 W $\sim 20.5$。考虑到最近在计算速度方面的改进，多波段PCAT显示出应用于大规模调查的前景，并且是多仪器观测数据联合分析的合理框架。
Simulation-Based Inference for Probabilistic Galaxy Detection and Deblending
University of Maryland, College Park（马里兰大学帕克分校）；University of Michigan（密歇根大学）；Leinweber Institute for Theoretical Physics, University of Michigan（密歇根大学莱因韦伯理论物理研究所）；The Voleon Group（Voleon集团）；Carnegie Mellon University（卡内基梅隆大学）
Identifying Anomalous DESI Galaxy Spectra with a Variational Autoencoder
University College London（伦敦大学学院）；Carnegie Mellon University（卡内基梅隆大学）；Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室）；Boston University（波士顿大学）；Universidad de La Laguna (ULL)（拉古纳大学）；Instituto de Astrofísica de Canarias（加那利天体物理研究所）；University of Rochester（罗切斯特大学）；Università degli Studi di Milano（米兰大学）
DeepDISC-Euclid: Source Classification and Photometric Redshifts in Euclid Deep Field North With a Pixel-Level Deep Learning Approach
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign（美国国家超级计算应用中心，伊利诺伊大学厄巴纳-香槟分校）；Center for Artificial Intelligence Innovation, University of Illinois at Urbana-Champaign（人工智能创新中心，伊利诺伊大学厄巴纳-香槟分校）；Max-Planck-Institut für extraterrestrische Physik（马克斯·普朗克地外物理学研究所）；Instituto de Astrofísica de Canarias (IAC)（加那利天体物理研究所；拉帕尔马大学天体物理学系）；Departamento de Astrofísica, Universidad de La Laguna (ULL)（布里斯托大学物理学院）；School of Physics, HH Wills Physics Laboratory, University of Bristol
Signal recognition and background suppression by matched filters and neural networks for Tunka-Rex
Institute of Applied Physics ISU（伊尔库茨克国立大学应用物理研究所）；Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院）；University of Zürich（苏黎世大学）；Vrije Universiteit Brussel（布鲁塞尔自由大学）；Skobeltsyn Institute of Nuclear Physics MSU（莫斯科国立大学斯科别利岑核物理研究所）
通卡无线电分机（通卡-雷克斯）是一个数字天线阵列，测量宇宙射线空气簇射在30-80兆赫频带内的无线电发射。Tunka-Rex与西伯利亚的TAIGA实验位于同一地点，由63个天线组成，其中57个位于约1 km \textsuperscript{2}的密集仪器区。在目前的工作中，我们讨论了改进的信号重建适用于通卡雷克斯。在第一阶段，我们实现了匹配滤波使用平均信号作为模板。仿真研究表明，匹配滤波可以降低信号检测阈值，提高信号纯度。然而，匹配滤波的最大性能仅在白噪声的情况下才能实现，而实际上噪声由于不同的原因而不是完全随机的。为了识别噪声的隐藏特征并对其进行处理，我们决定使用具有自动编码器架构的卷积神经网络。将记录的轨迹作为输入，自动编码器返回去噪轨迹，即删除所有与信号无关的幅度。比较了标准信号重构方法、匹配滤波方法和自动编码方法，讨论了神经网络在降低宇宙线探测数字天线阵阈值方面的应用前景。
