University of Amsterdam is looking for two PhD students interested in investigating geometric structure in data (e.g., structure of signals and shapes), machine learning architectures (e.g., group convolutional neural networks, graph neural networks, and representation embeddings), and scientific problems (e.g., with symmetry/equivariance constraints). You will employ the obtained insights to develop machine learning techniques that enable learning-based engineering for new data types and problems, and improve upon existing approaches on a fundamental level.
The primary focus of this project is on the analysis of medical image data for its variety of complex data structures and problems that could benefit from a deep geometric understanding of the data and underlying anatomical structures (e.g. shapes, texture and context). Although this is a core focus, you are encouraged to explore and demonstrate the broader applicability of geometric machine learning beyond medical image analysis, e.g., in computer vision, physics, molecular sciences, and scientific computing.
You are going to carry out research under the supervision of dr Erik Bekkers at the Amsterdam Machine Learning Lab (AMLab) of the Informatics Institute.
What are you going to do?
You are expected to:
invent, evaluate, and describe novel algorithms for the analysis of data based on geometric methods;
present research results at international conferences, workshops, and journals;
become an active member of the research community and collaborate with other researchers, both within and outside the Informatics Institute;
pursue, complete, and defend a PhD thesis within the contract duration of four years;
assist in teaching activities, such as teaching labs and tutorials or supervising bachelor and master students.
Further details:
2 PhD Positions at University of Amsterdam