Capsule Networks (CapsNets), recently proposed by the Google Brain team, have superior learning capabilities in machine learning tasks, like image classification, compared to the traditional CNNs. However, CapsNets require extremely intense computations and are difficult to be deployed in their original form at the resource-constrained edge devices. This paper makes the first attempt to quantize CapsNet models, to enable their efficient edge implementations, by developing a specialized quantization framework for CapsNets. We evaluate our framework for several benchmarks. On a deep CapsNet model for the CIFAR10 dataset, the framework reduces the memory footprint by 6.2x, with only 0.15% accuracy loss. We will open-source our framework at https://git.io/JvDIF.

Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks / Marchisio, Alberto; Bussolino, Beatrice; Colucci, Alessio; Martina, Maurizio; Masera, Guido; Shafique, Muhammad. - ELETTRONICO. - 1:(2020), pp. 1-6. (Intervento presentato al convegno 2020 57th ACM/IEEE Design Automation Conference (DAC) tenutosi a San Francisco, CA, USA, USA nel 20-24 July 2020) [10.1109/DAC18072.2020.9218746].

Q-CapsNets: A Specialized Framework for Quantizing Capsule Networks

Bussolino, Beatrice;Martina, Maurizio;Masera, Guido;
2020

Abstract

Capsule Networks (CapsNets), recently proposed by the Google Brain team, have superior learning capabilities in machine learning tasks, like image classification, compared to the traditional CNNs. However, CapsNets require extremely intense computations and are difficult to be deployed in their original form at the resource-constrained edge devices. This paper makes the first attempt to quantize CapsNet models, to enable their efficient edge implementations, by developing a specialized quantization framework for CapsNets. We evaluate our framework for several benchmarks. On a deep CapsNet model for the CIFAR10 dataset, the framework reduces the memory footprint by 6.2x, with only 0.15% accuracy loss. We will open-source our framework at https://git.io/JvDIF.
2020
978-1-7281-1085-1
File in questo prodotto:
File Dimensione Formato  
095_3.pdf

non disponibili

Descrizione: Articolo principale
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 1.24 MB
Formato Adobe PDF
1.24 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Q-CapsNets.pdf

accesso aperto

Descrizione: Versione autore
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 790.12 kB
Formato Adobe PDF
790.12 kB Adobe PDF Visualizza/Apri
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/11583/2848255