On the second day of the conference, Thursday, March 26, Radu Prodan (University of Innsbruck, Austria) will deliver a keynote address entitled “Distribute and Learn 1 Billion”, scheduled from 9:00 a.m. to 10:00 a.m. The presentation will take place in the Nicolae Iorga Amphitheater.
Abstract. The digital world hosts approximately 6 billion Internet users, representing about three-quarters of the global population, with over 8.5 billion mobile edge device subscriptions as the primary access method. Bringing nearly everyone online generates over 400 billion gigabytes of daily data, overwhelming the processing and storage capacity of modern supercomputers with millions of processing cores, operating at over 1 billion gigaflops per second. The emergence of GPU accelerators triggered the modern AI boom, enabling deep neural networks with billions of neurons and parameters to outperform humans in specific, supervised, structured tasks, such as internet search, image recognition, or speech detection. Today, edge devices like IoT industrial sensors, smart cameras, networking routers, switches, gateways, wireless access points, and consumer retail devices like smartphones, laptops, and healthcare equipment equipped with powerful neural processing units exhibit peak performance up to trillions of operations per second for specialized AI inference, comparable to that of parallel computers 20 years ago. The presentation provides an outlook on the research activities of the endowed professorship in Edge AI at the University of Innsbruck for learning on large numbers of distributed, interconnected, memory-constrained devices. The research addresses real industrial needs of 10 local funding companies and three large multimillion Horizon Europe projects: Graph-Massivizer using sampling for compressing graph-structured data with billions of nodes and trillions of edges for training and inference in graph neural networks, DataPACT addressing AI compliance according to the EU regulations (with over 19,000 total legal acts in force) using federated knowledge distillation, and CoAgent, resporting AI pipelines and distributed small language models for reasoning.