04 Postdoctoral Fellowship at Umea University, Sweden

Umeå University is one of Sweden’s largest higher education institutions with over 37,000 students and about 4,700 employees. The University offers a diversity of high-quality education and world-leading research in several fields. Notably, the groundbreaking discovery of the CRISPR-Cas9 gene-editing tool, which was awarded the Nobel Prize in Chemistry, was made here. At Umeå University, everything is close. Our cohesive campuses make it easy to meet, work together and exchange knowledge, which promotes a dynamic and open culture.

The ongoing societal transformation and large green investments in northern Sweden create enormous opportunities and complex challenges. For Umeå University, conducting research about – and in the middle of – a society in transition is key. We also take pride in delivering education to enable regions to expand quickly and sustainably. In fact, the future is made here.

Are you interested in learning more? Read about Umeå university as a workplace

The project is a collaboration between clinical, translational and basic science researchers with experience in surgery, pathology, diabetes research and 3D high-resolution imaging of the pancreas. Principal investigator is Oskar Franklin, MD, PhD at the Dept of Diagnostics and Intervention, Surgery and the project is a collaboration with Professor Ulf Ahlgren at the Dept of Medical and Translational Biology. The main part of the work will be carried out in Prof. Ahlgrens lab (webpage). The project uses innovative methods developed for diabetes research onto human pancreatic cancer applications, to improve understandning of microenviromental changes and applications for early detection and treatment. The work involved hands-on laborative and analytic work on samples from patients that are treated for pancreatic diseases. There will be opportunities to study both pancreatic cancer and diabetes using these techniques and to develop novel protocols for improved imaging and analysis. There will be opportunities for national and international collaboration.

The rapid development of autonomous systems, connected devices, and distributed applications poses several challenges in dealing with petabytes of data in diverse resource-constrained environments. Federated machine learning (FML) is a collaborative learning solution to handle these problems without sharing data with centralised servers. However, several emerging threats target FML training, learning, and inference to fail or mislead models at early learning rounds, particularly backdoor and bitflip attack and defence strategies under-explored in FML. These results jeopardize achieving trustworthy performance for any downstream tasks. Therefore, this project envisions developing and validating attack and defence strategies in federated learning for limited and diverse non-iid (independent identically distributed) data under non-standard and adversarial settings, which are ideally suited for edge AI infrastructures. These goals can be achieved by inducing unique features in federated learning algorithms such as robust training, model restoration, trustworthy device selection, secure learning and aggregation, fault-tolerance against failures and attacks, multi-level regularisers, and massive device variations. The ambition is to validate them in classical non-standard settings and apply them to solutions for constraint environments (e.g., the Internet of Things (IoT) and robotic arms). Potentially, teaching up to a maximum of 20% can be included in the work tasks.

This postdoctoral position encompasses the TAIGA focus areas Social AI, led by Associate Professor Loïs Vanhée and Understanding (and Explaining) AI, led by Assistant Professor Dimitri Coelho Mollo.

The Social AI focus area is dedicated to the study of AI methods that can adapt to a given social context, for instance by replicating human-like behavior (e.g. social simulation, virtual characters) or by human-centric adaptation to psychosocial factors (e.g. learning assistants sensitive to classroom dynamics, personalized mental health therapeutic agents.) 

The focus area Understanding (and Explaining) AI is dedicated to the study of how AI systems, especially large artificial neural networks, can be understood and explained in ways suited to different audiences, from AI researchers, to decision-makers, to the general public. This focus area uses tools from current research in philosophy of science and social epistemology to pursue an interdisciplinary investigation of what is needed for understanding and explaining AI.