About me

I am a PhD student in the ANTIQUE team at ENS-PSL / Inria / CNRS, supervised by Prof. Caterina Urban. My work focuses on neural specifications for formal methods in explainable AI.

My research interests lie in theoretical machine learning, explainable AI, and optimization. Previously, I worked on simplifying post-trained ReLU neural networks for mixed-integer linear programming systems at OVGU, and on data-driven decision analysis for energy systems at EPFL.

I graduated from École Polytechnique with an MSc&T in Artificial Intelligence & Advanced Visual Computing (AiViC, MX23).

Get in touch ↗·Download my CV ↗

Research interests

• Explainable AI & optimization
• Machine learning theory
• Computer vision & NLP
• Physics-informed neural networks & constraint learning

Publications & presentations

2026 / Preprint

Optimization with Region-Reduced ReLU Neural Networks

C. Plate, C. Ganzer, M. Hahn, A. Klimek, H. Liu, S. Sager, K. Sundmacher, and H. Wilhelm

arXiv:2609.31380, 2026

Read on arXiv ↗

2025 / Conference paper

On the role of artificial intelligence in feature oriented multi-criteria decision analysis

H. Liu, Y. Zhao, and F. Maréchal

ESCAPE 35, 2025

EPFL Infoscience ↗

All publications & presentations →

Beyond research

I enjoy basketball, playing the saxophone, and dragon boating.

Languages: English, Chinese, German (basic), and French (basic).