Oral Presentation

Peculiar Velocity Reconstruction from Simulations and Observations Using Deep Learning Algorithms

Author(s): Yuyu Wang (Ohio University)

Presenter: Yuyu Wang (Ohio University)

We implemented the Deep Learning Algorithm to develop a new model for reconstructing the 3D peculiar velocity field from the redshift distributions of halos and galaxies, simplifying the complex calculations in conventional velocity reconstruction methods. Utilizing convolutional neural networks, our model offers more accurate predictions with adaptability to realistic observational conditions, including the redshift space distortions, halo mass thresholds, and observational selection effects. We demonstrated the model’s efficacy by applying it to the Sloan Digital Sky Survey (SDSS) Data Release 7 for observational 3D peculiar velocity reconstructions.