Postdoctoral researcher in machine learning

Halmstad University

Work at a University where different perspectives meet!

Halmstad University adds value, drives innovation and prepares people and society for the future. Since the beginning in 1983, innovation and collaboration with society have characterised the University's education and research. The research is internationally reputable and is largely conducted in a multidisciplinary manner within the University's two focus areas: Health Innovation and Smart Cities and Communities. The University has a wide range of education with many popular study programmes. The campus is modern and well-equipped, and is situated close to both public transportation and the city center.

The School of Information Technology

Halmstad University’s School of Information Technology (ITE) is a renowned multi-disciplinary institution with around 155 employees from 20 different countries. ITE is internationally recognised for its applied research and collaborative initiatives, focusing on smart technology and its practical applications.

Within ITE, our students and researchers engage in diverse areas of study, including electronics, AI, information-driven healthcare, autonomous vehicles, social robotics, and digital design. We offer a comprehensive range of educational programmes, ranging from undergraduate to doctoral levels, as well as professional development opportunities.

Research is conducted within the University’s research programmes, especially Information Driven Care (IDC), Re-Imagining Future Smart Living – beyond the Living Lab (REBEL), Learning in a Digitalised Society (LeaDS) and the Future Industry Research Programme (FIRP).

ITE is home to Leap for Life, an innovation centre for information-driven care, as well as the Electronics Centre in Halmstad (ECH), a collaborative space for electronic development.

For more information about the School of Information Technology: https://hh.se/ite-en

More information about the School of Information Technology: hh.se/ite-en

More information about working at Halmstad University: https://hh.se/english/about-the-university/vacant-positions.html

Description

We seek a passionate researcher to join the CAISR group, working on cutting-edge machine-learning solutions for real-world industrial challenges. This role offers the opportunity to conduct independent and collaborative research leveraging tabular and time-series data, the most prevalent data type in industrial settings. Our focus includes advanced techniques like Meta-Learning, Self-Supervised Learning, and Foundational Models to push the boundaries of AI-driven industrial solutions.

The postdoc will carry out research, both independently and in collaboration with other members, in one or more cutting-edge AI/ML topics, including data mining, context-aware systems, knowledge-based intelligent systems, representation learning, meta-learning, transfer learning, multi-task, self-supervised and weakly-supervised learning, federated learning, anomaly detection, synthetic data generation, graph neural networks, evaluation, and more. Strength in data streams for industrial fault predictions will be a particular asset.

An example challenge addressed across several CAISR projects is how to jointly learn data representations that are useful for multiple tasks, both from data and expert knowledge, allowing for autonomous adaptation to a specific task. It includes physics-informed (and knowledge-engineered in other ways) machine learning, aware systems research, and autonomous knowledge creation. One challenge is extracting general features suitable for more than one task, often in self-supervised or semi-supervised settings.

How to best do AI/ML-based knowledge creation from largely unlabelled industrial data that leads to more efficient industry operations is an important research question for this position. This necessarily includes making sense of all available data, i.e., instead of striving for the unachievable ideal of perfect data that would correspond to well-controlled lab experiments, we learn from the industrial data that exists “in the wild”: either is being collected today, or is realistic to collect in the near future. The “largely unlabelled” phrasing means not only the ratio, i.e., the large amount of available data being unlabelled. It also, or maybe even primarily, means that many concepts of interest lack labels. While the vast majority of recent progress in AI/ML focuses on the supervised learning paradigm, we primarily work with what one could call “semi-unlabelled data.” Most of the data collection in the industry is done for purposes other than analytics, and thus, AI/ML uses data only tangentially relevant to the task at hand. It typically describes the operation of the equipment, or financial transactions and other business processes, but on-target labels are virtually non-existent. Some information can be used as labels, but even those are rare and unreliable. Most often, those labels come from auxiliary sources, and one can consider them “proxy” information, i.e., their meaning does not correspond to the things we are truly after but is associated (to different degrees) with them.

The recruited person can be involved, up to 20% of work time, in Bachelor’s and Master’s level courses such as Artificial Intelligence, Learning Systems, Data Mining, Applied Data Mining, and Deep Learning, including a newly started “Applied Artificial Intelligence” BSc program, graduate professional development program (second-cycle courses targeted at the business sector), supervision of thesis projects, outreach activities, and more.

Principal duties

The duties will be:

• Conduct machine learning research on time-series and tabular data for industrial applications, using advanced techniques like Meta-Learning, Self-Supervised Learning, and Large Models to improve efficiency and innovation.
• To contribute to academic publications and conference papers 
• To contribute to successful collaboration with our partners, including fulfillment of project goals and deliverables

Qualifications
The applicant must hold a doctoral degree in Artificial Intelligence/Data Mining/Machine Learning/Information Technology or related fields. The applicant needs to demonstrate a strong research profile in the fields related to topics of interest for CAISR research environment, including recent activities with high impact. The scientific production is expected to be published in high-quality, peer-reviewed research journals and conferences. Documented experience from innovation, research and development in an industrial environment is also a strong merit. The applicant should share the value that diversity and equality among researchers and teachers brings higher quality to research and education. 

Appointment to a post-doctoral position requires that the applicant has a PhD, or an international degree deemed equivalent to a PhD, within the subject of the position. The certificate, proving the qualification requirement is met, must be received before the employment decision is made. Primarily should those who have graduated no more than three years before the last application date, be considered. Under special circumstances, leave due to illness, parental leave, or similar circumstances, the doctoral degree can have been completed earlier.

For appointment as a postdoctoral researcher, the following assessment criteria will be applied:

• Ability to conduct research of high international quality in computer vision, robotics and artificial intelligence
• Documented experience from research in collaboration with industrial partners and/or interdisciplinary teams
• Ability to conduct high-quality teaching and to develop courses at different levels
• Experience in supervision of bachelor/master students
• Ability to attract external funding
• Dynamism, curiosity, independence, creativity, and good teamwork
• Willingness to address opportunities and challenges within AI, machine learning and data mining

Salary
Salary is to be settled by negotiation. The application should include a statement of the salary level required by the candidate.

Application
Applications should be sent via Halmstad University's recruitment system Varbi (see link on this page).

How to design your application

For further information, please contact Stefan Byttner (stefan.byttner@hh.se).

General Information
We value the qualities that gender balance and diversity bring to our organization. We therefore welcome applicants with different backgrounds, gender, functionality and, not least, life experience.
Read more about Halmstad University at http://hh.se/english/discover/discoverhalmstaduniversity.9285.html