Ce Sui

whusuice@gmail.com.

my_pic.jpg

Johns Hopkins University, MD, USA

I am a researcher working on cosmology, statistical inference, and machine learning. I obtained my PhD from the Department of Astronomy at Tsinghua University, China, under the advice of Prof. Yi Mao. During my PhD, I spent two years at IAP Paris and JHU in the USA, working with Prof. Ben Wandelt. I am currently a short-term research scientist at Johns Hopkins University, and will be joining Ohio State University later this year as a CASPER Fellow and CCAPP affiliated fellow.

Research

My research sits at the intersection of machine learning, statistical methods, and astronomy, using simulation-based inference, information theory, and generative models to tackle key data challenges in cosmology. I’m especially drawn to statistics as a rigorous mathematical language for expressing our intuitions about data — combined with modern ML tools, it becomes a powerful framework for tackling problems across science.

A major theme of my work is extracting and quantifying information from cosmic fields — from quantifying the information content of summary statistics, to extracting complementary information beyond existing summaries, to learning local information via score matching to enable Fisher forecasts without explicit likelihoods. I’ve also been exploring forward modeling for cosmological analysis, including a symbolic emulator for the matter power spectrum, and applying these methods in collaborative projects on galaxy clustering from the BOSS survey with the Learning the Universe Collaboration, and on 21cm signal inference, where I recently led a team in the SKA Data Challenge. Looking ahead, I hope to push these information-theoretic and generative approaches further, particularly toward robust forward modeling and field-level inference in cosmology.

Community & Service

I organized the Data Science Group at THU-DOA (2021–2023), exploring topics at the intersection of machine learning and astrophysics — see our repository. Since 2025, I’ve also been organizing the ML × Astro/Cosmo Group Meeting at Johns Hopkins University.

Hobbies

I’m a passionate gamer with a love for both video games and card games. Beyond playing, I’m also interested in game development and have been learning Unity3D to create my own games. One day, I hope to build a “universe” of my own—one that I actually understand :)

selected publications

2026

  1. How to Evaluate the Sufficiency and Complementarity of Summary Statistics for Cosmic Fields: An Information-theoretic Perspective
    Ce Sui, Yi Mao, Xiaosheng Zhao, and 2 more authors
    \apj, Feb 2026

2025

  1. SYREN-NEW: Precise formulae for the linear and nonlinear matter power spectra with massive neutrinos and dynamical dark energy
    Ce Sui, Deaglan J. Bartlett, Shivam Pandey, and 3 more authors
    Astronomy & Astrophysics, May 2025
  2. Fisher Score Matching for Simulation-Based Forecasting and Inference
    Ce Sui, Shivam Pandey, and Benjamin D. Wandelt
    arXiv e-prints, Jul 2025

2024

  1. Hybrid Summary Statistics
    T. Lucas Makinen, Ce Sui, Benjamin D. Wandelt, and 2 more authors
    Jul 2024