2021
DOI: 10.3847/1538-4357/abf3bb
|View full text |Cite
|
Sign up to set email alerts
|

Cosmic Velocity Field Reconstruction Using AI

Abstract: We develop a deep-learning technique to infer the nonlinear velocity field from the dark matter density field. The deep-learning architecture we use is a “U-net” style convolutional neural network, which consists of 15 convolution layers and 2 deconvolution layers. This setup maps the three-dimensional density field of 323 voxels to the three-dimensional velocity or momentum fields of 203 voxels. Through the analysis of the dark matter simulation with a resolution of 2h −1 Mpc, we find that t… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
15
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
4

Relationship

1
8

Authors

Journals

citations
Cited by 20 publications
(16 citation statements)
references
References 81 publications
0
15
0
Order By: Relevance
“…To our knowledge, observationally, accurate velocity measurements remain a challenge and are still fraught with problems, leading to errors that are difficult to eliminate. However, recently with new high-precision data, such as LSST (Ivezić et al 2019), DESI (Aghamousa et al 2016), CSST (Gong et al 2019), Euclid (Laureijs et al 2011;Amendola et al 2018), and advanced techniques it is possible to reliably reconstruct the cosmic velocity fields, e.g., a deep learning technique to infer the non-linear velocity field from the dark matter density field (Wu et al 2021), and a new Bayesian-based framework to infer the full three dimensional velocity field from observed distances and spectroscopic galaxies (Lavaux 2016). In addition, the presence of vorticity would leave observable effects, e.g., on redshift space distortions and the alignment of halo spins (Laigle et al 2014).…”
Section: Conclusion and Discussionmentioning
confidence: 99%
“…To our knowledge, observationally, accurate velocity measurements remain a challenge and are still fraught with problems, leading to errors that are difficult to eliminate. However, recently with new high-precision data, such as LSST (Ivezić et al 2019), DESI (Aghamousa et al 2016), CSST (Gong et al 2019), Euclid (Laureijs et al 2011;Amendola et al 2018), and advanced techniques it is possible to reliably reconstruct the cosmic velocity fields, e.g., a deep learning technique to infer the non-linear velocity field from the dark matter density field (Wu et al 2021), and a new Bayesian-based framework to infer the full three dimensional velocity field from observed distances and spectroscopic galaxies (Lavaux 2016). In addition, the presence of vorticity would leave observable effects, e.g., on redshift space distortions and the alignment of halo spins (Laigle et al 2014).…”
Section: Conclusion and Discussionmentioning
confidence: 99%
“…Since the data obtained from many cosmological observations can be processed as two-and threedimensional images, CNNs are well-suited to several aspects of cosmological simulations and data analysis. For example, some current CNN applications in cosmology include producing full-sky CMB simulations [72], identification of HII regions in reionization [73], analysis of dark matter substructure [74,75], and cosmic velocity field reconstruction [76].…”
Section: Resunet-cmb Architecture and Methodsmentioning
confidence: 99%
“…Motivated by the neural network model (Wu et al 2021), we use a modified UNet neural network architecture for model construction. The architecture of our neural network and its components are shown in Fig.…”
Section: Neural Network Modelmentioning
confidence: 99%