Foundations of AI & ML
Overview of theme
Research in our FAM division focuses on developing the theoretical and methodological foundations of Artificial Intelligence and Machine Learning. This involves topics such as Learning Theory, GenAI, Game Theory (Cooperative, non-Cooperative and Evolutionary), (Multi-agent) Reinforcement Learning, Computational Social Choice, Multi-agent Systems, Bandit theory, Monte Carlo methods, Asymptotics, Finite-sample guarantees, Generalisation and Regret bounds, Robustness, Misspecification, Generalised Bayesian Inference, Quantum Information and Learning Theory, Causality, Complexity of Learning Algorithms, analysis and methods for Federated Learning, Unlearning, Physics-informed Machine Learning, Parsimony, and Adaptation of LLMs and Diffusions.
Our key research areas include
| 1. Learning Efficiency & Complexity | 2. Robustness, Fairness, and Privacy |
| 3. Multiagent Systems, RL & Game Theory | 4. Generative AI & Alignment |
| 5. Causality & Causal Inference | 6. Quantum Computing and Learning Theory |
| 7. Stochastic Processes & Diffusions | 8. Physics-informed ML |
Publications and Projects
Our Publications list provides details of our published papers in books, journals and conferences.
We are involved in many diverse research projects funded by several external bodies such as UKRI, EPSRC, Unilink Ltd., Lloyds Register Foundation, The Alan Turing Institute, etc.
Foundations of AI & ML News
Foundations of AI & ML Events
No events to show.
PhD Applications
We welcome applications for PhD in the Foundations of AI starting in October 2026.