The portability of Convolutional Neural Networks (ConvNets) on the mobile edge of the Internet has proven extremely challenging. Embedded CPUs commonly adopted on portable devices were designed and optimized for different kinds of applications, hence they suffer high latency when dealing with the parallel workload of ConvNets. Reduction techniques playing at the algorithmic level are viable options to improve performance, e.g. topology optimization using alternative forms of convolution and arithmetic relaxation via fixed-point quantization. However, their efficacy is hardware sensitive. This paper provides an overview of these issues using as a case study an image classification task implemented through open-source resources, namely different architectures of MobileNet (vl), scaled, trained and quantized for the ImageNet dataset. In this work, we quantify the accuracy-performance trade-off on a commercial board hosting an ARM Cortex-A big. LITTLE system-on-chip. Experimental results reveal mismatches which arise from the hardware.

Inference on the Edge: Performance Analysis of an Image Classification Task Using Off-The-Shelf CPUs and Open-Source ConvNets / Peluso, V.; Rizzo, R. G.; Cipolletta, A.; Calimera, A.. - (2019), pp. 454-459. (Intervento presentato al convegno 6th International Conference on Social Networks Analysis, Management and Security, SNAMS 2019 tenutosi a esp nel 2019) [10.1109/SNAMS.2019.8931889].

Inference on the Edge: Performance Analysis of an Image Classification Task Using Off-The-Shelf CPUs and Open-Source ConvNets

Peluso V.;Rizzo R. G.;Cipolletta A.;Calimera A.
2019

Abstract

The portability of Convolutional Neural Networks (ConvNets) on the mobile edge of the Internet has proven extremely challenging. Embedded CPUs commonly adopted on portable devices were designed and optimized for different kinds of applications, hence they suffer high latency when dealing with the parallel workload of ConvNets. Reduction techniques playing at the algorithmic level are viable options to improve performance, e.g. topology optimization using alternative forms of convolution and arithmetic relaxation via fixed-point quantization. However, their efficacy is hardware sensitive. This paper provides an overview of these issues using as a case study an image classification task implemented through open-source resources, namely different architectures of MobileNet (vl), scaled, trained and quantized for the ImageNet dataset. In this work, we quantify the accuracy-performance trade-off on a commercial board hosting an ARM Cortex-A big. LITTLE system-on-chip. Experimental results reveal mismatches which arise from the hardware.
2019
978-1-7281-2946-4
File in questo prodotto:
File Dimensione Formato  
SNAMS19.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 335.98 kB
Formato Adobe PDF
335.98 kB Adobe PDF Visualizza/Apri
08931889_snam.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 925.25 kB
Formato Adobe PDF
925.25 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2816982