Devices for ubiquitous soils and temperature measurements.

Dit project ontdekken

AI Edge

Typically, a Deep Learning algorithm is trained on a powerful server with a big set of data. Then the trained network is loaded in an application which deals with real data. This step is called inference. However, inference does not have to be applied at the computer system which was used for training. Therefore, it is possible to apply deep learning on systems which are less powerful, more energy efficient, and less dependent on a network. This opens several opportunities.

Dit project ontdekken

Museum of future

Here in Belgium, we mainly use a traditional, old school way of teaching, where a teacher stands in front of the classroom and shares his knowledge. But this is far from efficient and very hard to use when we are trying to learn about futures. Also, there is so much to talk about that different opinions and perspectives are necessary.

Dit project ontdekken