I am a PhD student in Statistics and Machine Learning at the University of Copenhagen, supervised by Munir Hiabu. My research focuses on understanding machine learning models used for prediction, with the broader aim of developing a more rigorous foundation for Explainable AI. Some of the software projects I have contributed to can be found under Software. Before starting my PhD, I worked as a software engineer with experience in DevOps, Google Cloud, and CI/CD.
Jinyang Liu, Munir Hiabu (2026)
arXiv preprint arXiv:2605.31200. To appear in Proceedings of the 43rd International Conference on Machine Learning (ICML 2026)
We introduce Tensor Separation Learning, a glass-box regression model that learns sums of separable products of univariate functions. The model captures rich feature interactions while keeping its fitted components directly interpretable through first-order partial dependence functions.
Jinyang Liu, Tessa Steensgaard, Marvin N. Wright, Niklas Pfister, Munir Hiabu (2025)
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:39496-39534, 2025
We provide a novel and fast method for computing partial dependence functions for tree-based prediction models such as XGBoost and Random Forests. The implementation has since been integrated into the R-package glex.