Machine learning in statistical physics
Eligibility: UK/EU/International students with the required entry requirements
Award Details: No award (self funded)
Duration: Full Time - between 3 years and 3 years 6 months fixed term
Application deadline: Ongoing
Interview dates: TBC
Informal enquiries are essential before application; contact Martin Weigel to discuss this opportunity.
Coventry University has been voted ‘Modern University of the Year’ three times running by The Times/Sunday Times Good University Guide. Ranked in the UK’s top 15 (Guardian University Guide), we have a global reputation for high quality teaching and research with impact. Almost two-thirds (61%) of our research was judged ‘world leading’ or ‘internationally excellent’ in the Research Excellence Framework (REF) 2014. By joining the University’s Faculty of Engineering, Environment and Computing (EEC), you will benefit from state-of-the-art facilities and partnerships with some of the biggest names in industry, including Jaguar Land Rover, GE Aviation, Cummins and Intel.
Although in principle a well-established field, machine learning (ML) currently experiences a period of explosive expansion. This field is deeply rooted in statistical physics. In the present project, things are turned upside down and we investigate the use of methods of machine learning for studying phase transitions and critical phenomena in statistical physics. The student will use deep neural networks to quantitatively describe phase transitions in pure and disordered spin systems. Advanced techniques such as GANs will be employed to directly sample configurations of such systems. As a result, the student will be able to map out the utility of such methods for research in statistical physics and potentially open up exciting new avenues resulting from combining approaches from different disciplines.
- For the academic year 2018/19, any English student who is not part of a research council can borrow up to £25,000 to help cover the cost of their PhD tuition fees. Further details can be found here.
- The candidate will work with our extensive High-Performance Computing (HPC) facilities, including a GPU enabled 3000-core cluster
- Our research strategy is underpinned by a £250m investment in research and facilities
- Dedicated Doctoral College and Centre for Research Capability Development deliver high quality professional support for researchers, from PhD to Professor
- Free training: research career planning, managing your doctorate, research communication skills, research ethics, research impact, research integrity, research methods and research supervision
- Coventry is a member of the Doctoral Training Alliance (DTA), the largest multi-partner and only nationwide doctoral training initiative of its kind.
Successful applicants will have:
- A minimum of a 2:1 first degree in a relevant discipline/subject area with a minimum 60% mark in the Project element or equivalent with a minimum 60% overall module average.
- In the event of a first degree classification of less than 2:1, a Masters Degree in a relevant subject area will be considered as an equivalent. The Masters must have been attained with overall marks at minimum merit level (60%). In addition, the dissertation or equivalent element in the Masters must also have been attained with a mark at minimum merit level (60%), plus
- the potential to engage in innovative research and to complete the PhD within a three-year period of study
- a minimum of English language proficiency (IELTS overall minimum score of 7.0 with a minimum of 6.5 in each component)
- a BSc or MSc degree in Physics, Mathematics, or Computer Science
- some experience in statistical physics and/or machine learning
- programming skills in C/C++, Python, Fortran or other suitable languages