Learn more about techniques, applications and research challenges of uncertainty quantification which play a central role in reducing the impact of uncertainty in both optimisation and decision-making processes.

Examine how the problem of reconstructing a high-resolution colour image from burst of low-resolution raw pictures can addressed with a hybrid algorithm that retains the interpretability of the Lucas-Kanade approach to image alignment and iterative solutions to inverse problems but can be implemented in a feed-forward neural network, with parameters learned end to end.

And find out how Neuromorphic Computing can address the need for low power devices that are capable of advanced intelligence while reducing the number of specialized silicon chips and integration complexity.

Christophe Aufrère, Faurecia; Andreea Danielescu, Accenture Labs; Eric Moulines, Centre for Applied Mathematics at Ecole Polytechnique; Jean Ponce, Inria (National Institute for Research in Digital Science and Technology); Tim Shea, Intel