[The 89th TrustML Young Scientist Seminar] Talk by Song Liu (University of Bristol) "High-Dimensional Differential Parameter Inference in Exponential Family using Time Score Matching"

Date and Time:
February 27, 2025: 10:30 - 11:30 (JST)
Venue: Online and Open Space at the RIKEN AIP Nihonbashi office
Open Space is available to AIP researchers only

Title: High-Dimensional Differential Parameter Inference in Exponential Family using Time Score Matching

Speaker: Song Liu (University of Bristol)

Abstract:
This paper addresses differential inference in time-varying parametric probabilistic models, like graphical models with changing structures. Instead of estimating a high-dimensional model at each time and inferring changes later, we directly learn the differential parameter, i.e., the time derivative of the parameter. The main idea is treating the time score function of an exponential family model as a linear model of the differential parameter for direct estimation. We use time score matching to estimate parameter derivatives. We prove the consistency of a regularized score matching objective and demonstrate the finite-sample normality of a debiased estimator in high-dimensional settings. Our methodology effectively infers differential structures in high-dimensional graphical models, verified on simulated and real-world datasets. This is a joint work with Daniel J. Williams, Leyang Wang, Qizhen Ying and Mladen Kolar.

Bio:
Song Liu is an associate professor at University of Bristol, UK, specializing in statistical machine learning research and his research focuses on inference for intractable models using statistical discrepancies, and his work has been published in leading machine learning conferences such as ICML and NeurIPS in recent years. Song completed his PhD in Tokyo Tech in 2014, under the supervision of Prof. Masashi Sugiyama. He then moved to the Institute of Statistical Mathematics to become a postdoc in 2015, supervised by Prof. Kenji Fukumizu. He joined University of Bristol in 2017 where he continues to work on machine learning research.