The Challenge of Multi-Operand Adders in CNNs on FPGAs: How Not to Solve It!

Abstract : Convolutional Neural Networks (CNNs) are computationally intensive algorithms that currently require dedicated hardware to be executed. In the case of FPGA-Based accelerators, we point-out in this work the challenge of Multi-Operand Adders (MOAs) and their high resource utilization in an FPGA implementation of a CNN. To address this challenge, two optimization strategies, that rely on time-multiplexing and approximate computing, are investigated. At first glance, the two strategies looked promising to reduce the footprint of a given architectural mapping, but when synthesized on the device, none of them gave the expected results. Experimental sections analyze the reasons of these unexpected results.
Type de document :
Communication dans un congrès
18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS '18), Jul 2018, Pythagorion, Greece. ACM Press, pp.157-160, 2018, SAMOS '18 Proceedings of the 18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation. 〈10.1145/3229631.3235024〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01902952
Contributeur : Francois Berry <>
Soumis le : vendredi 18 janvier 2019 - 11:04:46
Dernière modification le : jeudi 21 février 2019 - 15:35:36

Fichier

1807.00217.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Kamel Abdelouahab, Maxime Pelcat, François Berry. The Challenge of Multi-Operand Adders in CNNs on FPGAs: How Not to Solve It!. 18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS '18), Jul 2018, Pythagorion, Greece. ACM Press, pp.157-160, 2018, SAMOS '18 Proceedings of the 18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation. 〈10.1145/3229631.3235024〉. 〈hal-01902952〉

Partager

Métriques

Consultations de la notice

47

Téléchargements de fichiers

8