TRKOY

Hi there 👋. I’m Tony RuiKang OuYang (歐陽瑞康 in Chinese; SeoiHong AuYeung pronounced in Cantonese). I’m an first-year PhD in Machine Learning 🤖 at the Machine Learning Group, University of Cambridge, supervised by Prof. José Miguel Hernández-Lobato and fully funded by the EPSRA DLA scholarship.

Recently, I completed my MPhil in Machine Learning and Machine Intelligence from the University of Cambridge 🎉. My MPhil thesis focus on energy-based neural sampler for sampling from Boltzmann distribution, which is supervised by Prof. José Miguel Hernández-Lobato. Prior to that, I finished my BEng in Data Science in Harbin Institute of Technology, Shenzhen (HITsz) and spent a wonderful year visiting in the University of Oxford studying Mathematics and Statistics (fully-funded by HITsz).

My current research interests include generative models, energy-based models, sampling methods, and their interactions and also their applications to molecular sciences 🧬. Generally, I’m interested in probabilistic machine learning and AI4Science, especially developing powerful, efficient, and scalable methods that can be applied to physics ⚛️ and biochemistry 🧪.

I lead the MolSS Reading Group, which invites top researchers to share their works on machine learning for molecular sciences, for every two weeks. Save our website and join our Slack-channel to stay tuned 🚀

selected publications

ICML
A Diffusive Classification Loss for Learning Energy-based Generative Models

RuiKang OuYang^*, Louis Grenioux^*, and José Miguel Hernández-Lobato

In International Conference on Machine Learning, 2026

* arXiv Bib

*denotes equal contribution. Order is randomly assigned.
@inproceedings{ouyang2026diffusiveclassificationlosslearning, title = {A Diffusive Classification Loss for Learning Energy-based Generative Models}, author = {OuYang, RuiKang and Grenioux, Louis and Hernández-Lobato, José Miguel}, year = {2026}, eprint = {2601.21025}, booktitle = {International Conference on Machine Learning}, archiveprefix = {arXiv}, primaryclass = {stat.ML}, url = {https://arxiv.org/abs/2601.21025}, star = {\* denotes equal contribution. Order is randomly assigned.} }
ICML
Progressive Tempering Sampler with Diffusion

Severi Rissanen^*, RuiKang OuYang^*, Jiajun He, and 4 more authors

In International Conference on Machine Learning, 2025

* arXiv Bib Code

*denotes equal contribution. Order decided by the winner of rock-paper-scissors (3:0): Severi (✋✋👊) v.s. ME (👊👊✂️).
@inproceedings{rissanen2025progressive, title = {Progressive Tempering Sampler with Diffusion}, author = {Rissanen, Severi and OuYang, RuiKang and He, Jiajun and Chen, Wenlin and Heinonen, Markus and Solin, Arno and Hern{\'a}ndez-Lobato, Jos{\'e} Miguel}, booktitle = {International Conference on Machine Learning}, year = {2025}, organization = {PMLR}, star = {\* denotes equal contribution. Order decided by the winner of rock-paper-scissors (3:0): Severi (✋✋👊) v.s. ME (👊👊✂️).} }
TMLR
BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matching

RuiKang OuYang^*, Bo Qiang^*, and José Miguel Hernández-Lobato

In Transactions on Machine Learning Research, 2025

* arXiv Bib Code Poster

*denotes equal contribution. Also published in Workshop on Machine Learning and the Physical Sciences @ NeurIPS 2024; and Workshop on ML4Molecules @ ELLIS 2024 (Oral)
@inproceedings{ouyang2025bnemboltzmannsamplerbased, title = {BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matching}, author = {OuYang, RuiKang and Qiang, Bo and Hernández-Lobato, José Miguel}, year = {2025}, eprint = {2409.09787}, archiveprefix = {arXiv}, primaryclass = {cs.LG}, url = {https://arxiv.org/abs/2409.09787}, booktitle = {Transactions on Machine Learning Research}, star = {\* denotes equal contribution. Also published in Workshop on Machine Learning and the Physical Sciences @ NeurIPS 2024; and Workshop on ML4Molecules @ ELLIS 2024 (Oral)} }