PhD in Adaptive Multi-Objective Search in Expensive High-Dimensional Socio-Technical Systems (with Honda Research Institute Europe) The University of Manchester in United Kingdom

This fully funded PhD project at The University of Manchester, in collaboration with the Honda Research Institute Europe (HRI-EU), focuses on developing advanced machine learning methods for expensive, high-dimensional multi-objective optimisation. The research addresses complex socio-technical systems, such as energy distribution networks for electric vehicles, where decision spaces are vast and objectives like efficiency, fairness, explainability, user satisfaction, and environmental impact must be balanced.

The project aims to overcome the limitations of current rule-based approaches by creating statistics-driven and learning-based optimisation methods. Techniques such as Bayesian optimisation with Gaussian process surrogate models will be explored to efficiently navigate high-dimensional design spaces and make optimal decisions under limited evaluation budgets. The research will build on recent advances in expensive optimisation, multi-objective surrogate modelling, and adaptive variable subset optimisation, leveraging both local and global search strategies.

Key research objectives include: exploring high-dimensional search spaces using intelligent variable subset selection, integrating socially derived criteria (trust, explainability, fairness), and coping with expensive evaluations by exploiting correlations in the search landscape. The successful candidate will have opportunities for regular visits to HRI-EU and to join Honda’s international PhD research community.

Applicants should have a strong quantitative background in machine learning, statistics, optimisation, or applied mathematics. Experience with Bayesian optimisation or Gaussian processes is highly desirable. The project is part of the UKRI AI CDT in Decision Making for Complex Systems and offers a stipend of £31,000 for students eligible for home fee status. The start date is September 2026.

To apply, candidates must submit a complete application through the University of Manchester Application Portal, specifying the project title and supervisor names, and uploading all required documents (transcripts, CV, supporting statement, and referee contact details). The university values equality, diversity, and inclusion, and encourages applicants from all backgrounds. Flexible and part-time study options are available.