How CERN and Ceva are pioneering the future of Edge AI

Edge AI brings computing capabilities close to where the data is collected i.e. on the device itself, rather than requiring transfer to the cloud. This is essential in particle detectors where the huge amounts of data and the time in which this data needs to be processed, means using the cloud just isn’t feasible.

CERN and Ceva joined forces to find a high-performing and power-efficient edge AI solution tailored specifically to CERN’s needs. In the process they have both gained from the collaboration. CERN have created new compression algorithms which can be used in future experiments and Ceva have developed a new technology, which can be integrated into their products.

Thanks to our collaboration with CERN we were able to develop an innovative approach that enables the networks to run up to 15 times faster compared to 16 bit baseline models. Its enhancing network speed and reducing energy consumption by up to 90% while maintaining comparable accuracy

– Olya Sirkin, Senior Deep Learning Researcher at Ceva

It exposed me, as a scientist, to the industrial way of thinking on a full-time basis. While the focus remained on experimental physics, we needed to ensure our solutions were transferable to Ceva's needs. It is always good to know how your model design choices will affect performance on different computing architectures - a skill I learned during the collaboration, thanks to feedback from Ceva’s experts

– Adrian Alan Pol, employed to work for this project and placed at CERN, now working as an Applied Machine Learning Specialist at Thomson Reuters