Xiaosheng Zhao

Denoising Diffusion Model for astrophysics

Zhao, X., Ting, Y., Diao, K., & Mao, Y., 2023. Can Diffusion Model Conditionally Generate Astrophysical Images?

Denoising Diffusion Probablistic Model (DDPM)

In this work, we apply the DDPM generative model for conditional generation of astrophysical images.

Wavelet scattering tranform and simulation-based inference from a Lightcone

Zhao, X., Zuo, S., & Mao, Y. 2023. 3D ScatterNet: Inference from 21 cm Light-cones.

Zhao, X., Zuo, S., Mao, Y. & Wandelt, B. D. 2023. Simulation-based Inference of Reionization Parameters from 3D Tomographic 21 cm Light-cone Images -- II: Application of Solid Harmonic Wavelet Scattering Transform.

3D ScatterNet architecture

In this work, we combine a 3D CNN featurizer and normalizing flows for astrophysical parameter inference from 3D 21 cm lightcones.

21 cm power spectrum and parameter uncertainty with machine learning

Zhao, X., Mao, Y., & Wandelt, B. D. 2022. Implicit Likelihood Inference of Reionization Parameters from the 21 cm Power Spectrum.

An example of the 21 cm power spectrum

In this work, we apply the DELFI to the 21 cm power spectrum. Realistic effects including thermal noise and foreground avoidance are also applied to the mock observations from the HERA and SKA.

3D 21 cm lightcone and parameter uncertainty with machine learning

Zhao, X., Mao, Y., Cheng, C., & Wandelt, B. D. 2022. Simulation-Based Inference of Reionization Parameters From 3D Tomographic 21 cm Lightcone Images

An illustration of the lightcone datacube of 3D 21 cm images

In this work, We perform a Bayesian inference of the reionization parameters where the likelihood is implicitly defined through forward simulations using density estimation likelihood-free inference (DELFI), where we further adopt a trained 3D Convolutional Neural Network (CNN) to compress the 3D image data into informative summaries (DELFI-3D CNN).