Assistant Professor In Geometric Learning and Image Analysis

In the Geometric Learning and Differential Geometry Group (GLDG) within the Centre for Analysis, Scientific Computing and Applications (CASA) we invent, apply, and build new theory and methods in geometric AI for automated medical and industrial image analysis. The Geometric Deep learning methods developed in GLDG bridge the gap in image processing between machine learning and traditional PDE/geometry-based methods.

The focus in the group GLDG supported by VICI project 'Geometric Learning for Image Analysis' (R. Duits) is on PDE-based equivariant neural networks aiming for impact on industry and society. This requires close interplay between mathematical theory and practical requirements. With implementation running parallel to theoretical development, we ensure that our geometric learning is both well-founded in profound new mathematics and relevant for practical applications.

Key objectives of the position are:

• Develop high-performance implementations of geometric learning methods for modern hardware while connecting advanced scientific computing to applied and geometric analysis.
• Establish connections to industry via joint projects on geometric learning and image processing, while investing in in-house implementations to keep our research relevant for industry.
• Target and steer the direction of the theory by available hardware and performance. Advanced geometrical, topological theory and algorithms need to be developed.
• Be active in funding acquisition and set up a clear, personal, and visible research line that fits the overall activities and strategy in the GLDG group and CASA cluster.
• Open the black box of deep learning via geometric interpretability and explainable AI. Promising results are growing and allow for clear analysis by experts, but intuitive visualization and interpretation tools to industrial/health tech partners raise challenges to be tackled.  
• Design new equivariant PDE-based neural networks. Current methods in GLDG reveal drastic network reduction, and training data-reduction with better performances. Pursue a research direction in explainable and transferable geometric AI.  Analysis of topological transitions, semi-ring co-domain extensions, Lie group domain extensions require new models.
• Connect to the GLDG group, the CASA cluster, and the M&CS department at TU/e, and act in cross-divisional institutes such as EAISI, to guarantee a vibrant, well-embedded and well-funded research group with better network design and image analysis applications.

A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you:

• An 18-month contract as Assistant Professor which can become permanent after 12 months on the same employment level upon successful evaluation.
• The opportunity to focus your career on education, research or impact. You can develop your personal ambitions and objections within the strategic objectives of the university. 
• Salary in accordance with the Collective Labour Agreement for Dutch Universities, scale 11 (min. €4492, max. €6148 gross per month).
• A year-end bonus of 8.3% and annual vacation pay of 8%.
• A dedicated mentoring program to help you get to know the university and the Dutch (research) environment.
• High-quality training programs for academic leadership and teaching.
• An excellent technical infrastructure, on-campus children's day care and sports facilities.
• Partially paid parental leave and an allowance for commuting, working from home and internet costs.
• A Staff Immigration Team is available for international candidates, as are a tax compensation scheme (the 30% facility) and partner career support.