ACACES is an annual summer school organised by the HiPEAC project (High Performance Embedded Architecture and Compilation). Giovanni Ansaloni has given a four day series of lectures at the 2024 iteration of this prestigious event
Maintaining rapid growth in computing performance is key for tackling the societal challenges shaping Europe and assuring our global competitiveness in the future.
“Edge intelligence has fostered a revolution in the way information is created, analyzed and accessed, fundamentally impacting multiple fields ranging from health monitoring to augmented reality,” explains Dr. Giovanni Ansaloni, of EPFL’s Embedded Systems Laboratory. “Key to this success is the optimization of application energy/run-time/application footprints, especially in light of the explosion in size of machine learning models vis-à-vis the tightly constrained computing and energy resources of edge architectures.
“The resulting landscape calls for a careful tailoring of both ‘what’ should be computed, as well as ‘how’ such computation should take place.”
Last month, at the 20th International Summer School on Advanced Computer Architecture and Compilation for High-performance Embedded Systems, Dr. Ansaloni gave a course of lectures that dealt with this landscape, entitled: Hardware/software co-optimization for machine learning at the edge.
Dr. Ansaloni discussed opportunities in both features-based and ML-based approaches, harnessing domain-specific characteristics such as sparsity and robustness.
He explored different techniques for the design of energy-minimal dedicated computing systems, ranging from smart SIMD architectures, to the use of in-and near-memory computing accelerators.
Importantly, he also showcased how open development hardware/software frameworks to facilitate holistic explorations.
“It really has been an honour to present the work of the Embedded Systems Lab at the ACACES24 Summer School,” enthused Dr. Ansaloni. “There was a great spirit of participation, and it was organised flawlessly.”
Explore open source tools and datasets developed and made available by the Embedded Systems Laboratory.
