Formal Methods in AI Rigorous foundations for trustworthy intelligent systems

Group members

Current group members, alumni, and visitors working on formal methods, logic, and machine learning for safe intelligent systems.

Current members

Dr. Francesco Belardinelli
Senior Lecturer, Principal Investigator
Formal methods in AI; strategic reasoning; safe and trusted reinforcement learning.
Department of Computing, Imperial College London
Dr. Edwin Hamel-de le Court
Postdoctoral Research Associate
Project: Abstraction-based technique for safe reinforcement learning (EPSRC NIA, 2023-25).
Omar Adalat
PhD Student
Verifiable and safe multi-agent reinforcement learning (2025-).
Maksim Anisimov
PhD Student
Improving robustness of neural networks using causal structure (2023-).
Alex Goodall
PhD Student
Robust world-model learning and shielding for reinforcement learning (2022-).
Titus Buckworth
PhD Student
Neuro-symbolic methods for AI (2022-).
Matt MacDermott
PhD Student
Model checking agents that learn (2021-).

Past members

Aamal Hussain
PhD Graduate (2019-24)
Asymptotic convergence and performance of multi-agent Q-learning dynamics.
Francis Rhys Ward
PhD Graduate (2020-2025)
Formalising and Evaluating Deception in AI Agents.
Borja Gonzalez Leon
PhD Graduate (2019-24)
Systematic grounded language learning in safety-aware autonomous agents.
Daniel Donnelly
MRes in AI (2023-24)
Runtime verification for safe and verifiable AI.
Luka Corliss
MRes in AI (2023-24)
Neuro-symbolic policy learning and representation for interpretable, verifiable reinforcement learning.
Nikolay Topol
MRes in AI (2022-23)
Explainable reinforcement learning with causality.

Visitors & student projects

Terrence
Masters' Internship (Jan 2026-)
Topic
© 2025-2026 Formal Methods in AI