Mathematician Kees Vuik looks at deep learning, an AI technique to handle big data
Most people would associate artificial intelligence with the data highway, but it is on actual highways that deep learning, a technique to manage huge data flows, could see its first application. Kees Vuik, director of the institute for computational science and engineering at Delft University of Technology, is working on a project to improve the monitoring of traffic on Dutch streets.
“If we supply a deep-learning network with the data from the GPS trackers in just 6 per cent of the cars on the road—which can be perfectly anonymised—we can already make better predictions of congestion than all installed expensive hardware and infrastructure,” he tells Research Netherlands.