The SMARTHEP-NextGen School on Edge Inference, took place at CERN from September 23-27th. This comprehensive training program was crafted to equip participants with both theoretical knowledge and practical experience in the field of Edge Machine Learning, with a focus on applications at the LHC.
Our program included introductory lectures, hands-on tutorials and seminars on a wide array of topics, including:
· Model compression techniques (AMD, hls4ml)
· Efficient GPU inference (Nvidia)
· Efficient FPGA inference (AMD)
· Neuromorphic computing and spiking neural networks (IBM Research)
· Heterogeneous accelerated computing (HACC)
In addition, we had talks on real-time inference applications across various domains such as autonomous vehicles, satellites, gravitational wave physics, and high energy particle physics.
The program also featured contributed flash talks and poster sessions, providing an excellent platform for attendees to present their work and exchange ideas.
