Dependability of Alternative Computing Paradigms for Machine Learning: hype or hope?

Cristiana Bolchini (1) , Alberto Bosio (2) , Luca Cassano (1) , Bastien Deveautour (2) , Giorgio Di Natale (3) , Antonio Miele (1) , Ian O'Connor (2) , Elena Ioana Vatajelu (3)

1 POLIMI - Politecnico di Milano [Milan]
2 INL - CSH - INL - Conception de Systèmes Hétérogènes
3 TIMA-AMfoRS - Architectures and Methods for Resilient Systems

Today we observe amazing performance achieved by Machine Learning (ML); for specific tasks it even sur- passes human capabilities. Unfortunately, nothing comes for free: the hidden cost behind ML performance stems from its high complexity in terms of operations to be computed and the involved amount of data. For this reasons, custom Artificial Intelligence hardware accelerators based on alternative com- puting paradigms are attracting large interest. Such dedicated devices support the energy-hungry data movement, speed of computation, and memory resources that MLs require to realize their full potential. However, when ML is deployed on safety- /mission-critical applications, dependability becomes a concern. This paper presents the state of the art of custom Artificial Intelligence hardware architectures for ML, here Spiking and Convolutional Neural Networks, and shows the best practices to evaluate their dependability.

Domaines: Micro et nanotechnologies/Microélectronique

Cristiana Bolchini, Alberto Bosio, Luca Cassano, Bastien Deveautour, Giorgio Di Natale, et al.. Dependability of Alternative Computing Paradigms for Machine Learning: hype or hope?. IEEE International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS 2022), Apr 2022, Prague, Czech Republic. ⟨10.1109/DDECS54261.2022.9770138⟩. ⟨hal-03641285⟩