4.6 Article

Implementation of Binarized Neural Networks in All-Programmable System-on-Chip Platforms

Journal

ELECTRONICS
Volume 11, Issue 4, Pages -

Publisher

MDPI
DOI: 10.3390/electronics11040663

Keywords

All Programmable System-on-Chip; Binarized Neural Networks; Convolutional Neural Network; Field-Programmable Gate Array

Ask authors/readers for more resources

This paper proposes a novel approach for implementing Binarized Neural Networks (BNN) on embedded systems, which preserves low-level features and requires minimal operations through optimization of hardware structure and BNN topology. Experimental results on the CIFAR-10 dataset demonstrate that the proposed method outperforms current BNN designs in terms of efficiency and accuracy.
The Binarized Neural Network (BNN) is a Convolutional Neural Network (CNN) consisting of binary weights and activation rather than real-value weights. Smaller models are used, allowing for inference effectively on mobile or embedded devices with limited power and computing capabilities. Nevertheless, binarization results in lower-entropy feature maps and gradient vanishing, which leads to a loss in accuracy compared to real-value networks. Previous research has addressed these issues with various approaches. However, those approaches significantly increase the algorithm's time and space complexity, which puts a heavy burden on those embedded devices. Therefore, a novel approach for BNN implementation on embedded systems with multi-scale BNN topology is proposed in this paper, from two optimization perspectives: hardware structure and BNN topology, that retains more low-level features throughout the feed-forward process with few operations. Experiments on the CIFAR-10 dataset indicate that the proposed method outperforms a number of current BNN designs in terms of efficiency and accuracy. Additionally, the proposed BNN was implemented on the All Programmable System on Chip (APSoC) with 4.4 W power consumption using the hardware accelerator.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available