LEDIG STILLING VED HØGSKULEN PÅ VESTLANDET

PhD Research Fellow in Machine Learning for Complex Networks

Deadline: 15.11.2021

Western Norway University of Applied Sciences

With about 16,000 students, Western Norway University of Applied Sciences is one of the largest higher education institutions in Norway. A broad range of academic programmes are offered at Bachelor, Master and PhD levels, spread out on five campuses Førde, Sogndal, Bergen, Stord og Haugesund.


Our ambition is to build stronger and more solid academic and research environments that will interact nationally and internationally. The aim is to become a recognized actor on the international higher education arena. Increased international cooperation and engagement in externally funded projects will work towards this goal.


The Faculty of Engineering and Science has approximately 300 employees and approximately 3,200 students. The faculty has a broad educational offer at both bachelor's and master's level in engineering and science, as well as PhD education in computer technology. The faculty's activities are internationally based and take place in close collaboration with regional companies, clusters, health trusts and the public sector, including other institutions in the university and college sector. This applies to research, development, innovation and not least education with student projects at all levels.

The Department of Computer Science, Electrical Engineering and Mathematical Sciences has a vacancy for a PhD research fellow in machine learning for complex networks for a period of 3 years

About the position:

The PhD research fellow will be part of the PhD programme in Computer Science: Software Engineering, Sensor Networks and Engineering Computing, HVL Data Science Group, and HVL Ci2Lab. The research programme in Computer Science currently includes 20 professors and associate professors, 40 PhD and post-doctoral research fellows, and a large number of master’s students.

About the PhD project/ research environment:

About the PhD project

Many real-world complex systems ranging from living organisms and human society to electricity, transportation, and the internet of things operate through multiple types of interactions between their components to perform their emergent functions. Therefore, these complex systems can be treated as multi-layer networks or networks of networks.

Despite advancements in network science, the vast majority of studies only deal with simplified models that are undirected, non-temporal, and those network models rarely take into account causal relationships and nonlinear structures. Failing to address concurrent dynamic processes in complex multi-layer networks creates significant challenges when addressing estimation and forecasting problems in such complex networks. Moreover, nonlinear time-dependency often makes it challenging to forecast dynamical behaviors in such networks. Furthermore, data scientists are wrestling with the computational challenge of scaling these models considering the explosive increase in the number of network components in real-world problems.

This project will create a mathematical framework by building from various monolayer networks' data entry. First, major nodes will be identified in each layer individually using graph- or information-based metrics. Then, we will reduce the multi-layer network size with nodes with the highest causal impact. This project's outcomes will drastically decrease the size of a network model, allowing scaling the model with feasible computation.

For better understanding the scope of this project, you can see our research lab webpage (www.ci2lab.com). Also, we published some paper reading multi-network modeling approaches for infrastructure networks as follows. However, this project’s use case also includes IoT systems in addition to infrastructure networks.

1. Resilience Characterization for Multi-Layer Infrastructure Networks, IEEE Intelligent Transportation Systems Magazine, 2021

2. Multi-Network Vulnerability Causal Model for Infrastructure Co-Resilience, IEEE ACCESS, 2019

3. Causal Markov Elman Network for Load Forecasting in Multinetwork Systems, IEEE Transactions on Industrial Electronics, 2019.

Workplace: campus Bergen.

Qualifications:

  • a master's degree in mathematics, computer science, control engineering, information engineering, electrical engineering (or a related field) *.
  • practical programming skills (e.g. R, Python, etc.) is a requirement.
  • A solid background in machine learning, network theory, or graph theory will be considered an advantage.
  • Knowledge in any area of Internet of Things will be considered a plus.

Candidates who have submitted a master’s thesis (but who has not yet been awarded a master’s degree) may also qualify for the position provided that the master`s degree is awarded by four weeks after the application deadline.

In addition to the required educational and technical background, the following criteria will be evaluated:

  • competence and grades on completed course work, quality of the master's thesis (excellent grade, equivalent of grade B or better on the ECTS grading system),
  • publications (if any), research, teaching experience, and industrial (if any) experience.
  • Applicants must be proficient in both written and oral English.
  • Personal and relational qualities will be emphasized. Ambitions and potential will also count when evaluating the candidates.

The candidate must be diligent and display the ability to work independently, supplemented with regular guidance, and is expected to carry out high-quality research and to publish the results in international conferences and journals.

The PhD research fellow must enroll in the PhD programme in Computer Science: Software Engineering, Sensor Networks and Engineering Computing at Western Norway University of Applied Sciences and must meet the formal admission requirements for admission into the PhD programme.

The project is 3 years with a possibility of extending the study to 4 years by designating 25% of the 4-year period to duties such as teaching, development and administrative tasks. The employment period may be reduced if the successful applicant has held previous employment as a research fellow.

Application procedure:

Applications will be evaluated by an expert panel of three members.

Applicants are asked to submit their application and CV online. Please use the link “Apply for this job” (“Søk stillingen”).

The following documentation should be uploaded as an attachment to the online application:

  • A CV with a list of related course works (e.g. in machine learning, network theory, statistics, programming) plus a list of academic publications
  • BS and MS Transcripts
  • MS Thesis
  • One page Statement of Purpose (SOP) expanding why the candidate is a good fit for this position
  • Copies of not more than 2 selected academic publications.

Applicants should indicate which publications or parts of publications should be given special consideration in the evaluation. If the documents submitted are not in a Scandinavian language or in English, the applicants must submit certified translations of these. The transcripts must specify the topics, the course works, and the grades at the bachelor`s and master`s degree levels.

Applicant whose education is from another country than Norway, need to also attach a certified translation of the diploma and transcript of grades to English or a Scandinavian language, if the original is not in any of these languages. It is required that the applicant enclose a review from NOKUT whether the education (bachelor and master’s degree) is of a scope and level that corresponds to the level of a Norwegian master’s degree. Please see www.nokut.no/en for more information about NOKUT’s general recognition. This may take some time and we recommend you to apply as soon as you know you will apply for this position. If no answer within the application deadline, please enclose documentation from NOKUT that they have received your application.

Applicants should note that the evaluation will be based on the documentation submitted electronically via Jobbnorge within the submission deadline. The applicants are responsible for ensuring that all the documentation is submitted before the closing date. It is of utmost importance that all publications to be considered in the evaluation are uploaded as an attachment with the application, since these are sent electronically to the expert panel. Applications cannot be sent by e-mail or to individuals at the college.

Salary:

  • Good occupational pension, insurance and loan schemes from The Norwegian Public Service Pension Fund
  • Exciting academic environment with the possibility of competence enhancement and development
  • Opportunities for training within the working hours

Salaries will be offered at grade 54 in the Civil Service pay grade table scale. There is a compulsory 2 % deduction to the pension fund (see http:/http://www.spk.no for more information). The successful applicant must comply with the guidelines that apply to the position at any time.

General information:

The appointment will be made in accordance with the regulations for State employees Law in Norway ("Lov om statens ansatte)". Organizational changes and changes in the duties and responsibilities associated with the position must be expected.

State employment shall reflect the multiplicity of the population at large to the highest possible degree. Western Norway University of Applied Sciences Bergen has therefore adopted a personnel policy objective to ensure that we achieve a balanced age and gender composition and the recruitment of persons of various ethnic backgrounds.

Information about the applicant may be made public even though the applicant has requested not to be named in the list of applicants. The applicant will be notified if his/her request is not respected.

Applicants may be called in for an interview.

Contacts:

1) Professor Reza Arghandeh, phone: +47 55 58 71 95, e-mail: [email protected]

2) Professor Håvard Helstrup, Coordinator of PhD Program on Computer Science, phone: +47 55 58 75 61, e-mail: [email protected]

Apply for this job

Powered by Labrador CMS