Machine learning meets sequential Monte Carlo methods
The project develops computationally efficient statistical tools to express model uncertainty in sequential prediction tasks.
Start date1 April 2023
Funding sourceEPSRC NPL iCASE studentship
- UK/International tuition fees
- Enhanced UKRI stipend at £20,668 p.a. (2022/23 rate)
- Research Training Support at £1,000 p.a.
- Personal Computer (provided by the Department).
This project lies on the intersection of modern machine learning techniques and sequential Monte Carlo methods. Sequential Monte Carlo (SMC) methods, e.g. particle filters, are a class of powerful simulation-based algorithms to utilise data uncertainty and generate model uncertainty.
Replacing the heuristic models in SMC by data-driven ones through the incorporation of machine learning will make them an extremely powerful tool in real-world applications including computer vision, finance, object tracking and robotics. This project will develop innovate statistical methods and apply the developed techniques in large-scale real-world datasets.
Start dates possible: April 2023, July 2023.
Related linksEUSIPCO 2022 Special Session on Machine Learning for Sequential Monte Carlo Met…
The application is rolling-based with no fixed submission deadline until the position is filled. Early applications are strongly encouraged for early PhD start.
The PhD student will be based at the Nature Inspired Computing and Engineering (NICE) research group in the Department of Computer Science at the University of Surrey. The student will also benefit from ample computing and research resources from Centre for Vision, Speech and Signal Processing (CVSSP) and the Surrey Institute for People-Centred AI.
A Bachelor’s degree or above in computer science, electrical engineering, statistics, mathematics, physics or similar (a first class or the equivalent from an overseas university).
Past research experience with prior publications is preferable although not essential.
English languages requirements
IELTS minimum 6.5 overall with minimum 6.0 in each component, or equivalent.