This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
Ponte: Represent Totally Binary Neural Network Toward Efficiency
by
Jia Xu
Jia Xu 1,2,3,
Han Pu
Han Pu 1,2 and
Dong Wang
Dong Wang 1,2,*
1
Institute of Information Science, Beijing Jiaotong University, Beijing 100044, China
2
Beijing Key Laboratory of Advanced Information Science and Network Technology, Beijing 100044, China
3
Intel Flex, Beijing 100091, China
*
Author to whom correspondence should be addressed.
Sensors 2024, 24(20), 6726; https://doi.org/10.3390/s24206726 (registering DOI)
Submission received: 21 August 2024
/
Revised: 12 October 2024
/
Accepted: 13 October 2024
/
Published: 19 October 2024
Abstract
In the quest for computational efficiency, binary neural networks (BNNs) have emerged as a promising paradigm, offering significant reductions in memory footprint and computational latency. In traditional BNN implementation, the first and last layers are typically full-precision, which causes higher logic usage in field-programmable gate array (FPGA) implementation. To solve these issues, we introduce a novel approach named Ponte (Represent Totally Binary Neural Network Toward Efficiency) that extends the binarization process to the first and last layers of BNNs. We challenge the convention by proposing a fully binary layer replacement that mitigates the computational overhead without compromising accuracy. Our method leverages a unique encoding technique, Ponte::encoding, and a channel duplication strategy, Ponte::dispatch, and Ponte::sharing, to address the non-linearity and capacity constraints posed by binary layers. Surprisingly, all of them are back-propagation-supported, which allows our work to be implemented in the last layer through extensive experimentation on benchmark datasets, including CIFAR-10 and ImageNet. We demonstrate that Ponte not only preserves the integrity of input data but also enhances the representational capacity of BNNs. The proposed architecture achieves comparable, if not superior, performance metrics while significantly reducing the computational demands, thereby marking a step forward in the practical deployment of BNNs in resource-constrained environments.
Share and Cite
MDPI and ACS Style
Xu, J.; Pu, H.; Wang, D.
Ponte: Represent Totally Binary Neural Network Toward Efficiency. Sensors 2024, 24, 6726.
https://doi.org/10.3390/s24206726
AMA Style
Xu J, Pu H, Wang D.
Ponte: Represent Totally Binary Neural Network Toward Efficiency. Sensors. 2024; 24(20):6726.
https://doi.org/10.3390/s24206726
Chicago/Turabian Style
Xu, Jia, Han Pu, and Dong Wang.
2024. "Ponte: Represent Totally Binary Neural Network Toward Efficiency" Sensors 24, no. 20: 6726.
https://doi.org/10.3390/s24206726
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article metric data becomes available approximately 24 hours after publication online.