PhD student in Computational Science with focus on High-Performance Computing

Project description

A tensor network is a flexible mathematical framework for efficiently approximating and computing with high-dimensional tensors. Tensor networks appear in a broad variety of scientific applications, ranging from the study of quantum many-body systems to autonomous driving and to explainable machine learning models. Computations with tensor networks present two main challenges: 1) The order in which nodes are contracted with one another affects both time and space requirements, and finding the optimal order is an NP complete problem. 2) In many applications, the tensors exhibit sparsity and properties that must be exploited for the contractions to become feasible. High-performance parallel libraries for the computation of tensor networks are still vastly missing. This project aims to develop novel algorithms, data structures, and software for tensor network algorithms in various applications.

Admission requirements 

The general admission requirements for doctoral studies are completion of the second-cycle studies, or completed course requirements of at least 240 ECTS credits, of which at least 60 ECTS credits are at second-cycle level, or an equivalent education from abroad. 

To fulfill the specific entry requirements for doctoral studies computational science and engineering, you are required to have completed at least 90 credits in computational science and engineering courses, of which at least 30 credits shall have been acquired at second-cycle level. You who otherwise have acquired skills that are deemed equivalent are also eligible.

Because of the interdisciplinary nature of the project, which aims to connect applications in physics with the domains of high-performance computing, numerical linear algebra, and tensor computations, the ideal candidate has knowledge of physics and its numerical methods, both discrete and continuous mathematics, as well as applied mathematics and parallel programming. Furthermore, the candidate needs to be proactive and independent in the identification of topics relevant to the conducted research, and in the acquisition of new knowledge.

Candidates must demonstrate very good knowledge in programming languages and in parallel programming, specifically for scientific computing applications. They must be skilled programmers (C/C++, Python, and Matlab), familiar with different computing paradigms (shared-memory and distributed-memory), and with the related libraries (OpenMP, MPI). The research will be carried out in collaboration with international researchers from different disciplines, so good collaborative skills are required, and you must be able to speak and write in English fluently.

The most important assessment criteria for the selection are scientific expertise and skills, suitability for the project, and documented ability to work in groups and to independently carry out work tasks.

About the position

The position provides you with the opportunity to pursue PhD studies in Computational Science for four years, with the goal of achieving the degree of Doctor in Computational Science and Engineering. While the position is mainly devoted to PhD studies (at least 80% of the time), it may include up to 20% department service (usually teaching). If so, the total time for the position is extended accordingly, resulting in a maximum of five years.

The procedure for recruitment for the position is in accordance with the Higher Education Ordinance (chapter 12, 2§) and the decision regarding the position cannot be appealed.

The expected starting date is September 1, 2024 or as otherwise agreed.