Erik Bekkers

Assistant professor
Informatics Institute
University of Amsterdam


Personal page   Google scholar   Github   Twitter  

I am a researcher in the field of geometric machine learning. In my work I focus on the mathematical foundations of deep learning and choose to work on topics that directly contribute to solving core problems in medical image analysis, aiming for generic solutions with a wide application scope. In the past I developed image analysis algorithms based on sub-Riemannian geometry in the Lie group SE(2) using the same mathematical techniques that are used to model the perceptual behavior of our visual system. Such brain inspired mathematics are not only relevant in image processing, but also more generally in machine learning where the use of symmetries and geometric structure provides powerful tools for robust and efficient representation learning and dealing with context. I currently work on group convolutional neural networks and related geometric machine learning techniques.