Project description
Third-cycle subject: Computer Science
The School of Electrical Engineering and Computer Science at KTH are looking for a doctoral student interested in Artificial Intelligence, Machine Learning, and Computer Vision that will work in a newly funded project financed by WASP (Wallenberg AI, Autonomous Systems and Software Program) which offers a graduate school with research visits, partner universities, and visiting lecturers. The aim of the research project is to develop more efficient vision transformer models. Specifically, we are interested in methods that allow transformers to scale up to operate on high resolution images, an essential task for medical image analysis. We are interested in developing more data-efficient training methods for vision transformers which reduce their reliance on large pretraining datasets. The project will include aspects of interpretability – providing insights into how vision transformers make decisions – and applications to medical image analysis.
Supervision
The doctoral student will be supervised by Associate Professor Kevin Smith (https://www.kth.se/profile/ksmith) whose research focuses on computer vision and biomedical image analysis. The research will be conducted at the Science for Life Laboratory (www.scilifelab.se) as part of the Division of Computational Science and Technology https://www.kth.se/cs/cst/division-of-computational-science-and-technology-1.779024 in the EECS school.