Hardware-Efficient Stochastic Binary CNN Architectures for Near-Sensor Computing - Archive ouverte HAL Access content directly
Journal Articles Frontiers in Neuroscience Year : 2022

Hardware-Efficient Stochastic Binary CNN Architectures for Near-Sensor Computing

, , (1) ,
1
Vivek Parmar
  • Function : Author
Bogdan Penkovsky
  • Function : Author
Manan Suri
  • Function : Author
  • PersonId : 1189850

Abstract

With recent advances in the field of artificial intelligence (AI) such as binarized neural networks (BNNs), a wide variety of vision applications with energy-optimized implementations have become possible at the edge. Such networks have the first layer implemented with high precision, which poses a challenge in deploying a uniform hardware mapping for the network implementation. Stochastic computing can allow conversion of such high-precision computations to a sequence of binarized operations while maintaining equivalent accuracy. In this work, we propose a fully binarized hardware-friendly computation engine based on stochastic computing as a proof of concept for vision applications involving multi-channel inputs. Stochastic sampling is performed by sampling from a non-uniform (normal) distribution based on analog hardware sources. We first validate the benefits of the proposed pipeline on the CIFAR-10 dataset. To further demonstrate its application for real-world scenarios, we present a case-study of microscopy image diagnostics for pathogen detection. We then evaluate benefits of implementing such a pipeline using OxRAM-based circuits for stochastic sampling as well as in-memory computing-based binarized multiplication. The proposed implementation is about 1,000 times more energy efficient compared to conventional floating-precision-based digital implementations, with memory savings of a factor of 45.
Fichier principal
Vignette du fichier
fnins-15-781786.pdf (4 Mo) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

hal-03861128 , version 1 (19-11-2022)

Identifiers

Cite

Vivek Parmar, Bogdan Penkovsky, Damien Querlioz, Manan Suri. Hardware-Efficient Stochastic Binary CNN Architectures for Near-Sensor Computing. Frontiers in Neuroscience, 2022, 15, ⟨10.3389/fnins.2021.781786⟩. ⟨hal-03861128⟩
0 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More