University of Aberdeen

Probabilistic
Machine Learning
Group

Advancing fundamental AI research in a lost paradise.

Probabilistic AI
in Scotland

The Probabilistic Machine Learning (PML) group, part of the Dept. of computing science at the University of Aberdeen, was founded by Dr. Yongchao Huang in Aug. 2023.

We are the first dedicated group of this kind at Aberdeen, focusing on fundamental research at the intersection of applied probability, computational statistics, and modern machine learning.

In the era of generative AI, our expertise in probabilistic, statistical and mathematical methods, along with industry experiences, uniquely position us to contribute to both theoretical understanding and practical applications of AI.

First of Its Kind

The inaugural probabilistic ML research group at the University of Aberdeen, establishing a new center of excellence.

Decade of Expertise

Combined experience spanning Bayesian methods, sampling algorithms, and probabilistic programming since 2015/2016.

Growing Community

The group currently hosts 3 PhDs and 2 MRes students, building & expanding a vibrant research community.

Our Focus

Probabilistic & Statistical Methods

Methodological innovations to machine learning through Bayesian inference, MCMC, variational inference, flows, GPs, physical and diffusion processes.

Uncertainties MCMC VI Flows Gaussian process Diffusion PPL

Mathematical Foundations of AI

Physics, mechanics and dynamics of learning, dynamical systems, numerical methods, and computing.

Learning Theory Dynamics Neural ODEs

Applications

Applying PML to real-world challenges in languages (e.g. SLM, uncertainties), vision, automation, and scientific computing.

NLP Vision Robotics Science AI

Researchers

YH

Dr. Yongchao Huang

Lecturer in Computing Science. Founder of the PML group & CPL Lab

Specializing in sampling and inference methods (MCMC, particle methods), Bayesian methods, computational and physical learning (CPL) approaches.

RH

RH

Researcher

Fuzzy logic systems.

HU

HU

Researcher

Quantumn ML.

Recent Publications

ArXiv

[2025] Probabilistic and reinforced mining of association rules

Yongchao Huang

AICS @ AAAI 2026

[2025] Training data membership inference via Gaussian process meta-modeling: a post-hoc analysis approach

Yongchao Huang, Pengfei Zhang and Shahzad Mumtaz

FPI @ ICLR 2025

[2024] Electrostatics-based particle sampling and approximate inference

Yongchao Huang

ArXiv

[2022] Classification via score-based generative modelling

Yongchao Huang

NeurIPS 2017 & 2020, ICML 2021, ICRA 2025

Applications: sensor network, smart building, cyber security, robotics

Y.H and collaborators

Join Our Group

We welcome inquiries from prospective PhD/Master students, postdoctoral researchers, and collaborators interested in probabilistic machine learning.

Location

Department of Computing Science
University of Aberdeen
Aberdeen, Scotland, UK

Follow

Y.H · GitHub