分类： 天文学 >> 天文学 提交时间： 2023-02-19
摘要： We introduce a new clustering algorithm, MulGuisin (MGS), that can find galaxy clusters using topological information from the galaxy distribution. This algorithm was first introduced in an LHC experiment as a Jet Finder software, which looks for particles that clump together in close proximity. The algorithm preferentially considers particles with high energies and merges them only when they are closer than a certain distance to create a jet. MGS shares some similarities with the minimum spanning tree (MST) since it provides both clustering and graph-based topology information. Also, similar to the density-based spatial clustering of applications with noise (DBSCAN), MGS uses the ranking or the local density of each particle to construct clustering. In this paper, we compare the performances of clustering algorithms using some controlled data and some realistic simulation data as well as the SDSS observation data, and we demonstrate that our new algorithm find clusters most efficiently and it defines galaxy clusters in a way that most closely resembles human vision.
分类： 天文学 >> 天文学 提交时间： 2023-02-19
Xiaoyuan Xiao
Yizhao Yang
Xiaolin Luo
Jiacheng Ding
Zhiqi Huang
Xin Wang
Yi Zheng
Cristiano G. Sabiu
Jaime Forero-Romero
Haitao Miao
Xiao-Dong Li
摘要： The mark weighted correlation function (MCF) $W(s,\mu)$ is a computationally efficient statistical measure which can probe clustering information beyond that of the conventional 2-point statistics. In this work, we extend the traditional mark weighted statistics by using powers of the density field gradient $|\nabla \rho/\rho|^\alpha$ as the weight, and use the angular dependence of the scale-averaged MCFs to constrain cosmological parameters. The analysis shows that the gradient based weighting scheme is statistically more powerful than the density based weighting scheme, while combining the two schemes together is more powerful than separately using either of them. Utilising the density weighted or the gradient weighted MCFs with $\alpha=0.5,1$, we can strengthen the constraint on $\Omega_m$ by factors of 2 or 4, respectively, compared with the standard 2-point correlation function, while simultaneously using the MCFs of the two weighting schemes together can be $1.25$ times more statistically powerful than using the gradient weighting scheme alone. The mark weighted statistics may play an important role in cosmological analysis of future large-scale surveys. Many issues, including the possibility of using other types of weights, the influence of the bias on this statistics, as well as the usage of MCFs in the tomographic Alcock-Paczynski method, are worth further investigations.
分类： 天文学 >> 天文学 提交时间： 2023-02-19
Ziyong Wu
Zhenyu Zhang
Shuyang Pan
Haitao Miao
Xin Wang
Cristiano G. Sabiu
Jaime Forero-Romero
Yang Wang
Xiao-Dong Li
摘要： We develop a deep learning technique to infer the non-linear velocity field from the dark matter density field. The deep learning architecture we use is an "U-net" style convolutional neural network, which consists of 15 convolution layers and 2 deconvolution layers. This setup maps the 3-dimensional density field of $32^3$-voxels to the 3-dimensional velocity or momentum fields of $20^3$-voxels. Through the analysis of the dark matter simulation with a resolution of $2 {h^{-1}}{\rm Mpc}$, we find that the network can predict the the non-linearity, complexity and vorticity of the velocity and momentum fields, as well as the power spectra of their value, divergence and vorticity and its prediction accuracy reaches the range of $k\simeq1.4$ $h{\rm Mpc}^{-1}$ with a relative error ranging from 1% to $\lesssim$10%. A simple comparison shows that neural networks may have an overwhelming advantage over perturbation theory in the reconstruction of velocity or momentum fields.
分类： 天文学 >> 天文学 提交时间： 2023-02-19
摘要： The distribution of matter that is measured through galaxy redshift and peculiar velocity surveys can be harnessed to learn about the physics of dark matter, dark energy, and the nature of gravity. To improve our understanding of the matter of the Universe, we can reconstruct the full density and velocity fields from the galaxies that act as tracer particles. We use a convolutional neural network, a V-net, trained on numerical simulations of structure formation to reconstruct the density and velocity fields. We find that, with detailed tuning of the loss function, the V-net could produce better fits to the density field in the high-density and low-density regions, and improved predictions for the amplitudes of the velocities. We also find that the redshift-space distortions of the galaxy catalogue do not significantly contaminate the reconstructed real-space density and velocity field. We estimate the velocity field $\beta$ parameter by comparing the peculiar velocities of mock galaxy catalogues to the reconstructed velocity fields, and find the estimated $\beta$ values agree with the fiducial value at the 68\% confidence level.
点击量
待评议
点击量
下载量
评论
