Dr Song Liu
BEng (Suzhou University), MSc (Bristol), Doctor of Engineering (Tokyo Tech)
Current positions
Associate Professor in Data Sciences and A.I.
School of Mathematics
Contact
Press and media
Many of our academics speak to the media as experts in their field of research. If you are a journalist, please contact the University’s Media and PR Team:
Biography
I do not update my publications here. Please see my publications on my website:
http://allmodelsarewrong.net
http://allmodelsarewrong.net
Projects and supervisions
Research projects
Using Machine Learning to Correct Probe Skew in High-frequency Electrical Loss Measurements
Principal Investigator
Role
Co-Investigator
Managing organisational unit
School of Electrical, Electronic and Mechanical EngineeringDates
20/11/2023 to 31/05/2024
Joint workshop Between JGI and ISM: High Dimensional and Bayesian Inference toward Quantifying Real-World Uncertainties
Principal Investigator
Description
This is the joint workshop between Jean Golding Institute (JGI) and the Institute of Statistical Mathematics (ISM) on High dimensional and Bayesian statistics. This workshop is the first of a…Managing organisational unit
Department of Computer ScienceDates
28/05/2019 to 31/05/2019
Publications
Recent publications
02/06/2025Guiding Time-Varying Generative Models by Natural Gradients on Exponential Family Manifold
Proceedings of the 41st Conference on Uncertainty in Artificial Intelligence
High-Dimensional Differential Parameter Inference in Exponential Family using Time Score Matching
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics
Score Matching with Missing Data
Proceedings of the 42nd International Conference on Machine Learning
Machine Learning Based Probe Skew Correction for High-frequency BH Loop Measurements
IEEE Transactions on Power Electronics
Approximate Stein Classes for Truncated Density Estimation
International Conference on Machine Learning (ICML 2023)