This paper presents a hybrid architecture for a neural network (NN) trained by a particle swarm optimization (PSO) algorithm. The NN is implemented on the hardware side while the PSO is executed by a processor on the software side. In addition, principal component analysis (PCA) is also applied to reduce correlated information. The PCA module is implemented in hardware by the SystemVerilog programming language to increase operating speed. Experimental results showed that the proposed architecture had been successfully implemented. In addition, the hardware-based NN trained by PSO (NN-PSO) program was faster than the software-based NN trained by the PSO program. The proposed NN-PSO with PCA also obtained better recognition rates than the NN-PSO without-PCA.
Tuan Linh DANG
Hanoi University of Science and Technology
Yukinobu HOSHINO
Kochi University of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Tuan Linh DANG, Yukinobu HOSHINO, "Hardware-Based Principal Component Analysis for Hybrid Neural Network Trained by Particle Swarm Optimization on a Chip" in IEICE TRANSACTIONS on Fundamentals,
vol. E102-A, no. 10, pp. 1374-1382, October 2019, doi: 10.1587/transfun.E102.A.1374.
Abstract: This paper presents a hybrid architecture for a neural network (NN) trained by a particle swarm optimization (PSO) algorithm. The NN is implemented on the hardware side while the PSO is executed by a processor on the software side. In addition, principal component analysis (PCA) is also applied to reduce correlated information. The PCA module is implemented in hardware by the SystemVerilog programming language to increase operating speed. Experimental results showed that the proposed architecture had been successfully implemented. In addition, the hardware-based NN trained by PSO (NN-PSO) program was faster than the software-based NN trained by the PSO program. The proposed NN-PSO with PCA also obtained better recognition rates than the NN-PSO without-PCA.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E102.A.1374/_p
Copy
@ARTICLE{e102-a_10_1374,
author={Tuan Linh DANG, Yukinobu HOSHINO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Hardware-Based Principal Component Analysis for Hybrid Neural Network Trained by Particle Swarm Optimization on a Chip},
year={2019},
volume={E102-A},
number={10},
pages={1374-1382},
abstract={This paper presents a hybrid architecture for a neural network (NN) trained by a particle swarm optimization (PSO) algorithm. The NN is implemented on the hardware side while the PSO is executed by a processor on the software side. In addition, principal component analysis (PCA) is also applied to reduce correlated information. The PCA module is implemented in hardware by the SystemVerilog programming language to increase operating speed. Experimental results showed that the proposed architecture had been successfully implemented. In addition, the hardware-based NN trained by PSO (NN-PSO) program was faster than the software-based NN trained by the PSO program. The proposed NN-PSO with PCA also obtained better recognition rates than the NN-PSO without-PCA.},
keywords={},
doi={10.1587/transfun.E102.A.1374},
ISSN={1745-1337},
month={October},}
Copy
TY - JOUR
TI - Hardware-Based Principal Component Analysis for Hybrid Neural Network Trained by Particle Swarm Optimization on a Chip
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1374
EP - 1382
AU - Tuan Linh DANG
AU - Yukinobu HOSHINO
PY - 2019
DO - 10.1587/transfun.E102.A.1374
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E102-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 2019
AB - This paper presents a hybrid architecture for a neural network (NN) trained by a particle swarm optimization (PSO) algorithm. The NN is implemented on the hardware side while the PSO is executed by a processor on the software side. In addition, principal component analysis (PCA) is also applied to reduce correlated information. The PCA module is implemented in hardware by the SystemVerilog programming language to increase operating speed. Experimental results showed that the proposed architecture had been successfully implemented. In addition, the hardware-based NN trained by PSO (NN-PSO) program was faster than the software-based NN trained by the PSO program. The proposed NN-PSO with PCA also obtained better recognition rates than the NN-PSO without-PCA.
ER -