-
will build on research that detects different types of uncertainty in deep neural networks, and we will connect this uncertainty to interactive data collection, e.g. in the form of a dialogue with the
-
modelling, data assimilation, and multi-scale neural network architectures applied to spatio-temporal data. The development of these methods is motivated by a concrete and important application: inferring gas
-
models and algorithms in Python, with documented experience in PyTorch. The applicant should be knowledgeable with neural networks and furthermore have a strong drive towards performing fundamental
-
-scale neural network models. While the developed methods will be broadly applicable, particular emphasis will be put on the problem of inferring gas dynamics in urban environments. Gas dynamics shape air