PhD candidate in Machine Learning at the edge of the Smart Grid

The University of Luxembourg seeks to hire outstanding researchers at its Interdisciplinary Centre for Security, Reliability and Trust(SnT), in the SERVAL team under Prof. Le Traon (http://wwwfr.uni.lu/snt/research/serval ). SnT is carrying out interdisciplinary research in secure, reliable and trustworthy ICT (Information and Communication Technologies) systems and services, often in collaboration with industrial, governmental or international partners.  SnT is active in several international research projects funded by national, European and international research programmes (e.g., FNR, Horizon Europe). For further information you may check: https://www.uni.lu/snt

We’re looking for people driven by excellence, excited about innovation, and looking to make a difference. If this sounds like you, you’ve come to the right place!

Edge and Fog computing will play a major role in speeding up the processes and reducing the delays in all data-oriented industries, as it opens up opportunities to perform computational tasks at the edge of the network that were not possible, or even conceivable, hitherto, such as Machine Learning (ML) training tasks. Empowering edge devices with lightweight, but yet optimized computational and storage capabilities is key, but also a challenge. A promising research direction consists in implementing ML model compression techniques such as model pruning, eventually combined with offloading strategies considering federated and/or multi-task learning-like strategies [Wa20, Mi20].

In this PhD thesis, the focus is on ML task offloading and resource allocation strategy under Fog-Cloud hybrid systems. More concretely, the research carried out in this thesis investigates how a given ML task can be divided into sub-tasks, and how those sub-tasks can be dynamically allocated between Cloud and Edge nodes under the multiple constraints of networking, security, computing power and energy consumption. This is known as a cost minimization problem under edge and cloud resource constraints, which is recognized as NP-hard [Ya18]. Federated and/or multi-task learning techniques will likely be considered in this respect, as they offer interesting features for task subdivision possibilities. The proposed ML task offloading strategy will be implemented and evaluated based on use case scenario defined in the (renewable) energy sector. The overall idea is to deploy the algorithm on nanogrid controllers that are limited in terms of computational capacity, memory and storage, which have to forecast future parameters such as anticipated energy demands and renewable generation outputs (e.g., predicted PV production).

Note that this PhD is granted by the “Fond National de la Recherche” (FNR) in Luxembourg, which officially starts in Sept. 2022 (4-year duration) and offers possibilities of collaboration with international research institutes such as the Rochester Institute of Technology in US.

[Wa20] Wang, X., Han, Y., Leung, V. C., Niyato, D., Yan, X., Chen, X. (2020). Convergence of edge computing and deep learning: A comprehensive survey. IEEE Communications Surveys & Tutorials, 22(2), 869-904.

[Mi20] Mishra, S., Sahoo, M. N., Bakshi, S., Rodrigues, J. J. (2020). Dynamic Resource Allocation in Fog-Cloud Hybrid Systems Using Multicriteria AHP Techniques. IEEE Internet of Things Journal, 7(9), 8993-9000.

[Ya18] Yang, L., Zhang, H., Li, M., Guo, J., Ji, H. (2018). Mobile edge computing empowered energy efficient task offloading in 5G. IEEE Transactions on Vehicular Technology, 67(7), 6398-6409.

The expected PhD candidate should have:

• A Computer Science background
• Expertise in Machine Learning and Optimization
• Good programming skills (python, Java…)
• Fluent written and verbal communication skills in English are mandatory
• Commitment, team working and a critical mind

• A stimulating learning environment. Here post-docs and professors outnumber PhD students. That translates into access and close collaborations with some of the brightest ICT researchers, giving you solid guidance
• Exciting infrastructures and unique labs. At SnT’s two campuses, our researchers can take a walk on the moon at the LunaLab, build a nanosatellite, or help make autonomous vehicles even better
• The right place for IMPACT. SnT researchers engage in demand-driven projects. Through our Partnership Programme, we work on projects with more than 45 industry partners
• Multiple funding sources for your ideas. The University supports researchers to acquire funding from national, European and private sources
• Competitive salary package. The University offers a 12 month-salary package, over six weeks of paid time off, health insurance and subsidised living and eating
• Be part of a multicultural family. At SnT we have more than 60 nationalities. Throughout the year, we organise team-building events, networking activities and more

But wait, there’s more!

Students can take advantage of several opportunities for growth and career development, from free language classes to career resources and extracurricular activities.

You will work in an exciting international environment (e.g., possibility to go to RIT in US during the PhD) and will have the opportunity to participate in the development of a newly created research centre.  

• Contract Type: Fixed Term Contract 36 Month (extendable up to 48 months if required)
• Work Hours: Full Time 40.0 Hours per Week
• Foreseen starting date: September 2022
• Employee and student status
• Location: Kirchberg
• Job Reference: UOL04686

The yearly gross salary for every PhD at the UL is EUR 37.101,36 (full time)

Applications in English should be submitted online and include:

• Curriculum Vitae (including your contact address, work experience, publications)
• Cover letter indicating your motivation and expected starting date (could depend on your ongoing formation)
• Contact information for 1 or 2 referees (incl., email + phone of your contacts)

All qualified individuals are encouraged to apply.

Early application is highly encouraged, as the applications will be processed upon reception. Please apply formally through the HR system. Applications by email will not be considered.

The University of Luxembourg embraces inclusion and diversity as key values. We are fully committed to removing any discriminatory barrier related to gender, and not only, in recruitment and career progression of our staff.

University of Luxembourg is an international research university with a distinctly multilingual and interdisciplinary character. The University was founded in 2003 and counts more than 6,700 students and more than 2,000 employees from around the world. The University’s faculties and interdisciplinary centres focus on research in the areas of Computer Science and ICT Security, Materials Science, European and International Law, Finance and Financial Innovation, Education, Contemporary and Digital History. In addition, the University focuses on cross-disciplinary research in the areas of Data Modelling and Simulation as well as Health and System Biomedicine. Times Higher Education ranks the University of Luxembourg #3 worldwide for its “international outlook,” #20 in the Young University Ranking 2021 and among the top 250 universities worldwide.

For inquiries please contact:

KUBLER Sylvain - [email protected]

LE TRAON Yves - [email protected]