Deep Neural Networks (DNN) have reached an outstanding accuracy in the past years, often going beyond human abilities. Nowadays, DNNs are widely used in many Artificial Intelligence (AI) applications such as computer vision, natural language processing and autonomous driving. However, these incredible performance come at a high computational cost, requiring complex hardware platforms. Therefore, the need for dedicated hardware accelerators able to drastically speed up the execution by preserving a low-power attitude arise. This paper presents innovative techniques able to tackle matrix sparsity in convolutional DNNs due to non-linear activation functions. Developed architectures allow to skip unnecessary operations, like zero multiplications, without sacrificing accuracy or throughput and improving the energy efficiency. Such improvement could enhance the performance of embedded limited-budget battery applications, where cost-effective hardware, accuracy and duration are critical to expanding the deployment of AI.
Low-Power Hardware Accelerator for Sparse Matrix Convolution in Deep Neural Network / Anzalone, Erik; Capra, Maurizio; Peloso, Riccardo; Martina, Maurizio; Masera, Guido. - ELETTRONICO. - 184:(2021), pp. 79-89. [10.1007/978-981-15-5093-5_8]
Titolo: | Low-Power Hardware Accelerator for Sparse Matrix Convolution in Deep Neural Network | |
Autori: | ||
Data di pubblicazione: | 2021 | |
Titolo del libro: | Progresses in Artificial Intelligence and Neural Systems | |
Serie: | ||
Abstract: | Deep Neural Networks (DNN) have reached an outstanding accuracy in the past years, often going be...yond human abilities. Nowadays, DNNs are widely used in many Artificial Intelligence (AI) applications such as computer vision, natural language processing and autonomous driving. However, these incredible performance come at a high computational cost, requiring complex hardware platforms. Therefore, the need for dedicated hardware accelerators able to drastically speed up the execution by preserving a low-power attitude arise. This paper presents innovative techniques able to tackle matrix sparsity in convolutional DNNs due to non-linear activation functions. Developed architectures allow to skip unnecessary operations, like zero multiplications, without sacrificing accuracy or throughput and improving the energy efficiency. Such improvement could enhance the performance of embedded limited-budget battery applications, where cost-effective hardware, accuracy and duration are critical to expanding the deployment of AI. | |
ISBN: | 978-981-15-5092-8 978-981-15-5093-5 | |
Appare nelle tipologie: | 2.1 Contributo in volume (Capitolo o Saggio) |
File in questo prodotto:
File | Descrizione | Tipologia | Licenza | |
---|---|---|---|---|
conv_accel.pdf | 2. Post-print / Author's Accepted Manuscript | PUBBLICO - Tutti i diritti riservati | Visibile a tuttiVisualizza/Apri | |
Wirn2019_Progresses in Artificial Intelligence and Neural Systems_Capra.pdf | 2a Post-print versione editoriale / Version of Record | Non Pubblico - Accesso privato/ristretto | Administrator Richiedi una copia |
http://hdl.handle.net/11583/2847352