Photometric redshifts from SDSS images with an interpretable deep capsule network
Photometric redshifts from SDSS images with an interpretable deep capsule network
复制标题
DOI：
10.1093/mnras/stac2105
复制
发表时间：
2021-12
影响因子：
4.8
通讯作者：
B. Dey;B. Andrews;J. Newman;Yao-Yuan Mao;M. Rau;R. Zhou
中科院分区：
文献类型：
--
作者：
B. Dey;B. Andrews;J. Newman;Yao-Yuan Mao;M. Rau;R. Zhou
宇宙学、星系演化和天文瞬变的研究，以及当前和下一代的宽视场成像调查，如鲁宾天文台的时空遗产调查（LSST），都严重依赖于光度红移的估计。胶囊网络是一种新型的神经网络结构，它比传统的卷积神经网络更适合于识别输入图像的形态特征。我们使用深度胶囊网络训练ugriz图像，光谱红移，银河动物园螺旋/椭圆分类的~ 400,000斯隆数字巡天（SDSS）星系进行光度红移估计。我们在需要更少的数据和更少的可训练参数的情况下，实现了与SDSS主星系样本样数据集（r≤17.8和zspec≤0.4）相当或优于当前方法的光度红移预测精度和灾难性异常值的一小部分。此外，我们的胶囊网络的决策更容易解释，因为胶囊作为图像的低维编码。当胶囊投影在二维流形上时，它们形成一个单一的红移序列，其中螺旋的分数在一个区域内显示出大致垂直于红移序列的梯度。我们在这个低维空间中扰动真实星系图像的编码，以创建合成的星系图像，这些图像显示了每个维度编码的图像属性（例如大小，方向和表面亮度）。我们还测量了星系属性（例如星等、颜色和恒星质量）与每个胶囊维度之间的相关性。我们在https://biprateep.github.io/encapZulate-1上公开发布我们的代码、估计的红移和其他目录。
Studies of cosmology, galaxy evolution, and astronomical transients with current and next-generation wide-field imaging surveys like the Rubin Observatory Legacy Survey of Space and Time (LSST) are all critically dependent on estimates of photometric redshifts. Capsule networks are a new type of neural network architecture that is better suited for identifying morphological features of the input images than traditional convolutional neural networks. We use a deep capsule network trained on ugriz images, spectroscopic redshifts, and Galaxy Zoo spiral/elliptical classifications of ∼400,000 Sloan Digital Sky Survey (SDSS) galaxies to do photometric redshift estimation. We achieve a photometric redshift prediction accuracy and a fraction of catastrophic outliers that are comparable to or better than current methods for SDSS main galaxy sample-like data sets (r ≤ 17.8 and zspec ≤ 0.4) while requiring less data and fewer trainable parameters. Furthermore, the decision-making of our capsule network is much more easily interpretable as capsules act as a low-dimensional encoding of the image. When the capsules are projected on a 2-dimensional manifold, they form a single redshift sequence with the fraction of spirals in a region exhibiting a gradient roughly perpendicular to the redshift sequence. We perturb encodings of real galaxy images in this low-dimensional space to create synthetic galaxy images that demonstrate the image properties (e.g. size, orientation, and surface brightness) encoded by each dimension. We also measure correlations between galaxy properties (e.g. magnitudes, colours, and stellar mass) and each capsule dimension. We publicly release our code, estimated redshifts, and additional catalogues at https://biprateep.github.io/encapZulate-1.