PhD in Logic, Learning, and Graph-Structured Data at Queen Mary University of London
Queen Mary University of London (QMUL) is offering a fully funded PhD position in the School of Electronic Engineering and Computer Science, focusing on bridging logic and learning for graph-structured data. The project aims to advance the integration of formal logic, verification, and automated reasoning with neural learning, addressing key challenges in artificial intelligence such as trust, safety, interpretability, and accountability. The research will explore how formal methods—including logic, model checking, and automated theorem proving—can enhance the reliability and transparency of neural models, with a central focus on graph-structured data, which is foundational in many AI applications and theoretical computer science.
Potential research directions include characterizing the expressive power of neural models using formal languages, verifying AI systems against symbolic specifications, extracting logical rules and structured explanations from trained models, and providing formal derivations to justify or constrain model predictions. While the project is primarily theoretical, there is an ambition to implement and evaluate developed methods on real datasets, potentially in collaboration with industrial partners. The outcomes are expected to contribute to the foundations of safe, explainable, and interpretable AI, a priority in both academia and industry.
Applicants should have a strong background in at least one of the following areas: first-order or modal logics, model checking and verification, logic programming, automated theorem proving, or automata theory. Experience with neural models or practical implementation is an advantage but not required. Intellectual curiosity, mathematical maturity, and motivation to work on deep problems are essential. Candidates must qualify as UK home students, with no restrictions on their stay in the UK and having been ordinarily resident in the UK for at least three years prior to the start of the studentship. English language certification is required for non-native speakers.
The position is fully funded, covering tuition fees at the home rate and providing a London stipend at QMUL rates (currently £21,874 per year for 2025/26, subject to confirmation for 2026/27) for three years. The application deadline is 26 January 2026, with the position starting in April or September 2026. Applicants should work with their prospective supervisor, Dr Przemysław Wałęga, and submit their application online, including a CV, cover letter, statement of research interest, two references, sample of written work, and English language certificate if required. For more information, visit the QMUL website or contact the supervisor or administrative staff.
