Hi there! My name is Xing (pronunciation; similar to “H’sing”). I am a Data Scientist at QuantCo, where my work involves building statistical models for insurance pricing.
Previously, I finished my PhD in Statistics at Imperial College London. I was in the EPSRC CDT in Modern Statistics and Statistical Machine Learning at Imperial and Oxford from 2020 to 2024, and I was advised by Professor Axel Gandy and Dr Andrew Duncan. I was also a visiting student to the Fundamentals of Statistical Machine Learning research group at UCL from autumn 2023 to summer 2024.
My research interests lie in the intersection of kernel methods and computational statistics. Specifically, I am exploring the applications of Kernelized Stein Discrepancy in various fields including hypothesis testing, particle-based inference and parameter estimation.
Education
PhD in Modern Statistics and Statistical Machine Learning, 2020-2024, Imperial College London
MASt (Part III) in Mathematical Statistics, 2019-2020, University of Cambridge
BSc in Mathematics with Statistics, 2016-2019, Imperial College London
Resources
Here is an unofficial LaTex poster template for maths/stats projects with a Imperial College theme. See the links therein for references.
Teaching
I am/was a Teaching Assistant for the following courses:
- Autumn 2023: M.Sc. in Statistics Orientation Week. Led by Dr Oliver Ratmann.
- Spring 2023: Mathematical Foundations of Machine Learning. Lectured by Dr Anastasia Borovykh.
- Spring 2022: Exploratory Data Analysis and Visualisation. Lectured by Dr James Martin.
- Autumn 2021: Applicable Maths. Lectured by Dr James Martin.
News
[03/2026] Our new preprint Data Fusion with Distributional Equivalence Test-then-pool is out! This is led by Linying Yang and in collaboration with Prof Robin Evans.
[03/2026] Our new preprint Kernel Tests of Equivalence is out! This is a joint work with Prof Axel Gandy.
[10/2025] Our paper On the Robustness of Kernel Goodness-of-Fit Tests, coauthored with Professor François-Xavier Briol, has been accepted to JMLR. Link.
[12/2024] I joined QuantCo as a full-time Data Scientist. Excited to explore the more applied side of machine learning in this role!
[12/2024] I successfully defended my PhD thesis! A huge thank-you to my examiners Dr. Nikolas Kantas and Prof. Chris Oates for the insightful discussions during the viva!
[08/2024] From 12th to 16th August 2024, I will attend the 11th Bernoulli-IMS World Congress in Probability and Statistics, and present our recent preprint On the Robustness of Kernel Goodness-of-Fit Tests, a joint work with Dr François-Xavier Briol. Come and join our session on Wednesday 14th at 11am if you are interested!
[08/2024] Our new preprint On the Robustness of Kernel Goodness-of-Fit Tests is out! This is a joint work with Dr François-Xavier Briol.
[12/2023] From 17th to 21st, I will attend the 2023 IMS International Conference on Statistics and Data Science (ICSDS) in Lisbon, where I will give a contributed talk on our COLT paper A High-dimensional Convergence Theorem for U-statistics with Applications to Kernel-based Testing.
[10/2023] For this academic year, I will be visiting the Fundamentals of Statistical Machine Learning research group, co-led by Dr François-Xavier Briol and Dr Jeremias Knoblauch. Looking forward to an engaging experience with the amazing researchers in the group!
[10/2023] Starting from October 2023, I will become an Enrichment Student at the Alan Turing Institute, where I will join the Turing’s research community for six months to broaden my research. Please do not hesitate to reach out if you are interested in collaboration!
[05/2023] Our paper A High-dimensional Convergence Theorem for U-statistics with Applications to Kernel-based Testing has been accepted by COLT 2023.
[04/2023] Our paper Using Perturbation to Improve Goodness-of-Fit Tests based on Kernelized Stein Discrepancy has been accepted by ICML 2023.
[09/2022] I finished my internship at Meta, where I worked on multi-task active learning methods for e-commerce.
[01/2022] Our paper Grassmann Stein Variational Gradient Descent has been accepted by AISTATS 2022.