Random spin systems from statistical mechanics have been successfully used to study critical behaviour at the interface of problems from mathematics, physics, biology and computer science.
The measure describing the spin system gives the probability of likely configurations. Phase transitions occur when tuning a parameter of the system, in the sense that there are several Gibbs measures compatible with the microscopic interactions. In recent years, there has been enormous activity in order to study phase transitions for spin systems on random graphs by rigorously characterizing the set of Gibbs measures. Applications of such problems range from computational complexity, coding theory, to machine learning.
This PhD project investigates statistical mechanics on random graphs and algorithms. The focus is on properties of the q-state Potts model, which is intimately related to random graph q-coloring and community detection problems.
As a PhD student your task is to:
- study relevant literature and learn the necessary mathematical techniques in order to tackle the problems mentioned above;
- present results on scientific conferences;
- attend research seminars, local at Technical University Eindhoven (TU/e) and the national Mark Kac seminar;
- publish articles in renowned scientific journals;
- participate in the activities of the project team involving PhD students and PostDocs;
- do modest teaching duties at TU/e.
The project is a joint project between Prof. Remco van der Hofstad (TU/e), Dr. Noela Mueller (TU/e) and Dr. Wioletta Ruszel (UU).