jason mclaren
phd student · school of engineering maths and technology
university of bristol
i am a phd student on the data-driven engineering and sciences program. working at the broad intersection of machine learning and engineering systems. in terms of application areas i am a generalist, although i have a particular interest in space physics/systems and robotics.
research interests
optimisation
gaussian processes, bayesian neural networks, and uncertainty quantification for engineering models.
structural health monitoring
data-driven approaches for damage detection, localisation, and prognosis in engineering structures.
physics-informed learning
embedding domain knowledge and physical constraints into machine learning architectures.
digital twins
creating adaptive computational models that evolve with their physical counterparts using streaming data.
transfer learning
population-based approaches for sharing information across heterogeneous engineering structures.
active learning
optimal experimental design and sensor placement for resource-efficient data acquisition.
featured projects
all projects →bayesian structural health monitoring
2025a framework for probabilistic damage detection in bridge structures using gaussian process models. the approach quantifies epistemic uncertainty in damage predictions, enabling risk-informed maintenance scheduling.