论文标题
使用机器学习中的回归技术估算星系和QSO的光度红移
Estimating the Photometric Redshifts of Galaxies and QSOs Using Regression Techniques in Machine Learning
论文作者
论文摘要
测量星系，恒星和类星体等宇宙学来源的距离在现代宇宙学中起着越来越重要的作用。获取光谱并因此将红移计算为距离指示器可以立即对这些对象进行分类。只要许多星系无法获得光谱观测，并且测量红移的过程是耗时的，并且对于大型样品而言，可以采用机器学习（ML）方法来确定来自不同特征的星系的红移，包括其光度颜色。在本文中，通过使用Sloan Digital Sky Survey（SDSS）目录中的通量幅度，我们开发了两种ML回归算法（决策树和随机森林）来估算以颜色指数为输入特征的红移。我们发现，随机森林算法可为红移预测产生最佳结果，当数据集仅限于带有z $ \ le $ 2的子集时，它将得到进一步改进，从而为标准化的标准偏差$ \ edline {Δz} _ {\ text _ {\ text {norm}}} = 0.005 $ and $ nastard deviation $ $ $ c} = 0.这项工作显示了使用ML方法确定遥远来源的光度红移的巨大潜力。
Measuring distances of cosmological sources such as galaxies, stars and quasars plays an increasingly critical role in modern cosmology. Obtaining the optical spectrum and consequently calculating the redshift as a distance indicator could instantly classify these objects. As long as spectroscopic observations are not available for many galaxies and the process of measuring the redshift is time-consuming and infeasible for large samples, machine learning (ML) approaches could be applied to determine the redshifts of galaxies from different features including their photometric colors. In this paper, by using the flux magnitudes from the Sloan Digital Sky Survey (SDSS) catalog, we develop two ML regression algorithms (Decision Tree and Random Forest) for estimating the redshifts taking color indices as input features. We find that the Random Forest algorithm produces the optimum result for the redshift prediction, and it will be further improved when the dataset is limited to a subset with z $\le$ 2 giving the normalised standard deviation $\overline{ΔZ}_{\text {norm}}=0.005$ and the standard deviation $σ_{Δz}=0.12$. This work shows a great potential of using the ML approach to determine the photometric redshifts of distant sources.
