Projects
Statistical / stochastic computing. KU Leuven
Although state-of-the-art in many typical machine-learning tasks, deep learning algorithms are very costly in terms of energy consumption, due to their large amount of required computations and huge model sizes. Because of this, deep learning applications on battery constrained wearables have only been possible through wireless connections with a resourceful cloud. This setup has several drawbacks. First, there are privacy concerns. ...
Model Reduction of Contact Problems in Flexible Multibody Dynamics KU Leuven
The research presented in this dissertation is concerned with the efficient solution of dynamic contact problems by means of distributed-parameter methods. Such problems are encountered in the computer-aided design of mechanical systems, where contact interactions are ubiquitous and often vital to the proper functioning of the system. With weight reduction and increased operating frequencies among the primary objectives of present-day ...
General Framework for Creative Artificial Intelligence KU Leuven
CIRCULATIONS AND IMAGINARIES: PHOTOGRAPHY AND MULTILITERACY IN THE POST-MODERN ERA KU Leuven
This interdisciplinary research at the crossroads of language sciences, art and history of photography, conducted in parallel at the University Sorbonne Nouvelle and the KU Leuven, will aim at showing how photography and language didactics, which emerged throughout the nineteenth century and share common political and economic stakes, have participated in the construction of representations of otherness. It will examine the role of images in ...
Birgittine Monasteries in the Low Countries as communities of learning and sites of knowledge transmission in a European context (c. 1440-1600). University of Antwerp
Rediscovering organized non-religious discourses in the United States and the United Kingdom, 1881-2019 KU Leuven
Non-religion studies constitute an emerging sub-field within the broader context of sociology of religion. For this group of scholars the question is no longer whether religion is or not on its way out, but rather what is progressively replacing it. However, the sudden mismatch between new theoretical interests and former goals of data collection makes the existing data largely sub-optimal for the tasks currently at hand. Indeed, for decades ...
Representation Learning for Sign Language Translation Using Linguistic and Knowledge-based Constraints KU Leuven
This PhD project is an integral part of the Horizon 2020 research project SignON, which unites 17 European partners in a research project that aims to facilitate the exchange of information among deaf and hard of hearing, and hearing individuals across Europe by developing automatic sign language translation tools. Recent machine learning methods based on neural transformer architectures have greatly improved the state-of-the art in natural ...