PhD: Physics-Informed Neural Networks: Imperial College London invites motivated candidates to apply for a PhD position in the Scalable Scientific Machine Learning Lab, supervised by Dr. Ben Moseley. The project focuses on developing advanced physics-informed neural networks (PINNs) to perform efficient and accurate multi-scale simulations across domains such as climate science, geophysics, and materials science. The successful candidate will join a highly interdisciplinary team working at the intersection of machine learning, applied mathematics, and high-performance computing.
PhD Scholarship Opportunity: Multi-Scale Simulation with Physics-Informed Neural Networks
Designation
PhD Student (Doctoral Researcher)
Research Area
- Multi-scale simulation
- Physics-informed neural networks (PINNs)
- Multi-GPU and high-performance computing
- Applied mathematics, computational physics, and scientific machine learning
- Applications in climate modelling, earthquake simulation, and biological/physical systems
Location
Scalable Scientific Machine Learning Lab
Department of Earth Science and Engineering
Imperial College London, UK
Eligibility/Qualification
Essential:
- Strong Master’s degree (or exceptional Bachelor’s with research portfolio) in applied mathematics, physics, computer science, engineering, or related fields
- Coursework in machine learning and/or applied mathematics
- Proficiency in programming (Python, C++, Julia, or Fortran)
Desirable:
- Experience in numerical modelling (finite difference, finite element, spectral methods, etc.)
- Understanding of scientific machine learning, especially PINNs
- Familiarity with deep learning frameworks (PyTorch, JAX/Equinox)
- Experience in HPC, GPU, or parallel computing
- Publications and/or relevant industry experience
Job Description
The PhD candidate will:
- Design and develop physics-informed neural networks for large-scale, multi-scale simulations
- Implement methods across multiple GPUs to enable scalable training
- Investigate algorithmic improvements (e.g., adaptive domain decomposition, random feature methods, linear solvers)
- Apply developed methods to real-world problems such as turbulent fluid simulations and regional earthquake modelling
- Collaborate with domain specialists and present research at high-impact conferences and journals
How to Apply
Interested candidates should send:
- CV (including education and research experience)
- Motivation letter (200–400 words) highlighting suitability for the project
- Optional: additional supporting materials
📧 Contact: Dr. Ben Moseley – b.moseley@imperial.ac.uk
For further details on the lab’s PhD application process, visit the Scalable Scientific Machine Learning Lab website.
For the official Imperial College PhD application process, refer to the Imperial PhD Admissions page.
Last Date to Apply
Applications are reviewed on a rolling basis.
Please check Imperial College scholarship deadlines for funding opportunities.
