Illustration of Benjamin L'Huillier running, working at a laptop, and playing music, against a cosmology blackboard

Benjamin L'Huillier

Statistical methods for cosmology

Timeline of Benjamin L'Huillier's career from 2007 to 2026: transition from engineering, PhD in Paris, KIAS, KASI, Yonsei, and Sejong University
Tap to enlarge

My science journey redux

I am an Assistant Professor in the Department of Physics and Astronomy at Sejong University in Seoul, South Korea. I develop statistical and computational methods (Gaussian processes, non-parametric reconstruction, Bayesian model selection, deconvolution) and use them to test the standard model of cosmology with data from DESI, supernovae and gravitational lensing.

My research is supported by the National Research Foundation of Korea. I explore the limits of the concordance ΛCDM model using N-body simulations—both within and beyond the standard model—and apply advanced statistical methods to test its assumptions. I was also a member of the DESI collaboration, contributing to the time domain and the cosmological simulations working groups.

Before my PhD, I trained as an engineer in signal and image processing at Supélec (now CentraleSupélec), and completed an MSc in fundamental and applied mathematics (Université Paul Verlaine, Metz) and an MSc in astrophysics (Université Paris Diderot). I completed my PhD at LUX, Paris Observatory and Paris Sorbonne University, working in the Galaxies and Cosmology group with Françoise Combes and Benoit Semelin, and as a member of the Momentum project. During that time, I also taught physics and data analysis as a teaching assistant at the Université Paris Cité.

Since 2012, I have been based in South Korea—a growing hub for research, in particular in astronomy and cosmology. I was a Research Fellow from 2012 to 2016 at the Korea Institute for Advanced Study (KIAS), working with Profs. Changbom Park and Juhan Kim in the Astrophysics and Cosmology group on cosmological structure formation using numerical simulations such as Horizon Run 4. I then joined the CosKASI group at the Korea Astronomy and Space Science Institute in Daejeon, working with Arman Shafieloo, before becoming Research Professor and W. C. Rufus Fellow working with James Jee at Yonsei University. Since 2021, I have also been an associate member of KIAS, and from 2022 to 2025 I was an associate member of the Center for Gravitational-Wave Universe at Seoul National University.

Methods

The same tools serve cosmology and any problem where an unknown function has to be inferred from noisy, indirect data.

Gaussian process regression

How the choice of mean function and the way hyperparameters are set bias GP reconstructions, and how to avoid it.

Hwang+ 2023, JCAP

See the papers

Non-parametric reconstruction

Reconstruct the expansion and growth histories directly from data, without assuming a model (iterative smoothing, Gaussian processes), then test models against them.

L’Huillier & Shafieloo 2017; Shafieloo+ 2018; L’Huillier+ 2018, 2019, 2020, 2025; Calderón+ 2022, 2023; Millard+ 2026

See all the papers

Model selection

Bayesian evidence compared with a frequentist approach based on iterative smoothing.

Koo+ 2021, 2022, JCAP

See the papers

Inverse problems

Recovering the matter power spectrum by Richardson–Lucy deconvolution of lensing data.

Dawn+ 2026; Jiang+ 2026

See the papers

Large-scale simulation

Cosmological N-body simulations: initial conditions and numerical accuracy, halo interactions and alignments, dark energy and modified gravity, counts-in-cells statistics.

Horizon Run 4 (Kim+ 2015); L’Huillier+ 2012, 2014, 2015, 2017; Uhlemann+ 2017, 2018; Hassani+ 2020

See all the papers

Ongoing

Optimal transport and machine learning to scale these reconstructions to Stage-IV data, with my students.

The group

News

Detection of an orthogonal alignment between parsec-scale AGN jets and their host galaxies

Egg-shaped galaxies may be aligned to the black holes at their hearts.

Read article

Where to find me

Address

Department of Physics and Astronomy, Sejong University

Gwangjin-gu, Neungdong-ro 209, Seoul 05006, South Korea

Email

benjamin@sejong.ac.kr