TechRxiv
20220226_The_Case_for_In-Memory_Inferencing_of_Quantized_CNNs_at_the_Edge.pdf (2.29 MB)
Download file

The Case for In-Memory Inferencing of Quantized CNNs at the Edge

Download (2.29 MB)
preprint
posted on 07.03.2022, 03:04 by Gabriel Falcao, João Dinis FerreiraJoão Dinis Ferreira
As artificial intelligence becomes a pervasive tool for the billions of IoT devices at the edge, the data movement bottleneck imposes severe limitations on these systems’ performance and autonomy. Processing-in-Memory emerges as a way to mitigate the data movement bottleneck while satisfying the stringent performance, energy efficiency, and accuracy requirements of edge imaging applications that rely on convolutional neural networks.

Funding

EXPL/EEI-HAC/1511/2021

Hardware Accelerated Deep LEarning framework

Fundação para a Ciência e Tecnologia

Find out more...

UIDB/EEA/50008/2020

History

Email Address of Submitting Author

gff@deec.uc.pt

ORCID of Submitting Author

0000-0001-9805-6747

Submitting Author's Institution

Instituto de Telecomunicações, Universidade de Coimbra

Submitting Author's Country

Portugal