TU Delft-s four Vici laureates are:
Theory and Inference of shortest and nearly shortest paths in large partially observed networks through learning their representations in non-Euclidean metric spaces.
Dr. M.A. Kitsak, Electrical Engineering, Mathematics and Computer Science (EEMCS)
Finding the shortest pathThe shortest path between the origin and destination node pair in a network is the shortest ordered sequence of nodes one needs to traverse through the network to get from origin to destination. The shortest path is an important concept in transportation, telecommunication, systems biology, and epidemic spreading. Finding a shortest path is nearly impossible with conventional graph-theoretic methods when the network is only partially observable. This project will develop a theory and computational inference algorithms for finding shortest paths using geometric network representations in non-Euclidean spaces. Project theory will be tested in practical settings for Internet routing and biomedicine.
Personal page
Project dAIta: Data Efficient AI Foundation Models
Dr. J.C. van Gemert, Electrical Engineering, Mathematics and Computer Science (EEMCS)
AI with less dataThe foundation of AI is built on top of huge, uncurated data. This data is predominantly controlled by a handful of private companies, what makes us dependent on them. Because of the size, this data is difficult to curate for bias, privacy, or otherwise sensitive data. I will make use of prior knowledge to reduce the amount of data needed to train AI: All knowledge that is built-in no longer has to be learned from data.
Personal pag
The Catalyst Dance: Understanding Ensemble Dynamics for Theory-Guided Catalyst Development (DYNACAT)
E.A. Pidko, Applied Sciences (AS)Taking the lead in the catalyst dance: teaching everyday metals choreography to power the future
In the intricate ’dance’ of chemical reactions, catalysts take the lead. However, the star performers in catalysis are often the noble metals that are most scarce and costly. In this project, scientists will teach abundant, everyday metals to dance like their more privileged noble metal counterparts. By combining advanced theory, experiments and automation, researchers will track, control, and optimize these catalysts, making green chemistry more attainable and efficient.
Personal page
Connecting quantum dots in three dimensions
Dr. ir. M. Veldhorst, QuTech (QT)
Connecting quantum dots in three dimensionsElectrons confined in arrays of quantum dots can be tuned at the individual level. The ambition of this project is to advance these arrays into the third dimension. These systems may lead to new emergent physics, be used for quantum simulations of real physics occurring in nature, and for quantum computer that have the promise of solving problems out of reach for classical computers.
Personal page
