We offer a position at a leading technical university that generates knowledge and skills for a sustainable future. You will have engaged and ambitious colleagues along with a creative, international and dynamic working environment. You will work in Gothenburg but with the close collaboration with the research team at Ericsson in Stockholm.

Information about the research and the project

The purpose of the project is to develop a scalable distributed machine learning paradigm (referred to as federated learning (FL) where the long convergence time, caused by model update/aggregation between the FL clients and the central server is mitigated. Current approaches prevent the FL from being applied for many emerging low-latency services, e.g., autonomous driving. The project will implement multiple threads in FL, referred to as multi-thread federated learning (MFL), to boost learning accuracy and efficiency.

The following activities will be performed in the frame of the project:
1. Multi-thread protocol design, where multiple threads are introduced in the training process. Each training round is composed of three elements, namely local training, model exchange, and model aggregation/update. The principle of the MFL is to interleave various elements in different training rounds, fully utilizing the local computation power to speed up the training process. In this activity, a communication protocol tailored for the MFL will be designed. New model aggregate/update functions will also be investigated, which allows the central server to coordinate the different threads and fully explore the benefits of the MFL. Moreover, other key aspects will be investigated, incl. robustness, scalability and energy efficiency.
2. Queueing model will be derived to analyze the training time and gain a fundamental understanding of the convergence time for the FL. In the cases where the analytical model will be too complex, the discrete event driven simulator will be developed to bound the delay performance for the general traffic patterns.
3. Use case specification and validation will be carried out through tight collaborations with Ericsson. Use cases that are highly aligned with Ericsson's solutions for 5G and beyond will be prioritized. Given a specified use case, the performance of the designed MFL will be validated.

Major responsibilities
Design of an appropriate multi-thread protocol. Deriving and validating of the related queuing models

Ph.D. degree in Electrical Engineering or Computer Science.
Knowledge within communication systems and machine learning algorithms.

Research experience in:
- queueing models,
- network architecture,
- resource allocation,
- distributed/federated machine learning algorithms,
- strong skills in using Python and Tensorflow,
- publication record on ML related work.

Contract terms
This researcher position is a full-time temporary employment for one year.

We offer
Chalmers offers a cultivating and inspiring working environment in the coastal city of Gothenburg
Read more about working at Chalmers and our benefits for employees.

Chalmers aims to actively improve our gender balance. We work broadly with equality projects, for example the GENIE Initiative on gender equality for excellence. Equality and diversity are substantial foundations in all activities at Chalmers.

Application procedure
The application should be marked with Ref 20220586 and written in English. The application should be sent electronically and be attached as PDF-files, as below. Maximum size for each file is 40 MB. Please note that the system does not support Zip files.

CV: (Please name the document: CV, Family name, Ref. number)
• CV
• Other, for example previous employments or leadership qualifications and positions of trust.
• Two references that we can contact.

Personal letter: (Please name the document as: Personal letter, Family name, Ref. number)
1-3 pages where you:
• Introduce yourself
• Describe your previous experience of relevance for the position (e.g. education, thesis work and, if applicable, any other research activities)
• Describe your future goals and future research focus

Other documents:
• Copies of bachelor and/or master’s thesis.
• Attested copies and transcripts of completed education, grades and other certificates, e.g. TOEFL test results.

Please use the button at the foot of the page to reach the application form. 

Application deadline: January 31st, 2023

For questions, please contact:
Professor Lena Wosinska, Electrical Engineering
Email: wosinska@chalmers.se

*** Chalmers declines to consider all offers of further announcement publishing or other types of support for the recruiting process in connection with this position. *** 

Chalmers University of Technology conducts research and education in engineering sciences, architecture, technology-related mathematical sciences, natural and nautical sciences, working in close collaboration with industry and society. The strategy for scientific excellence focuses on our six Areas of Advance; Energy, Health Engineering, Information and Communication Technology, Materials Science, Production and Transport. The aim is to make an active contribution to a sustainable future using the basic sciences as a foundation and innovation and entrepreneurship as the central driving forces. Chalmers has around 11,000 students and 3,000 employees. New knowledge and improved technology have characterised Chalmers since its foundation in 1829, completely in accordance with the will of William Chalmers and his motto: Avancez!