Accuracy Analysis of Node Activation Function Based on Hardware Implementation of Artificial Neural Network

被引：0

作者：

Jiang, Nan ^{[1
]}

Hou, Ligang ^{[1
]}

Guo, Jia ^{[1
]}

Zhang, Xinyi ^{[1
]}

Lv, Ang ^{[1
]}

机构：

[1] Beijing Univ Technol, VLSI & Syst Lab, Beijing, Peoples R China

来源：

2018 3RD IEEE INTERNATIONAL CONFERENCE ON INTEGRATED CIRCUITS AND MICROSYSTEMS (ICICM) | 2018年

关键词：

artificial neural network; VLSI; piecewise nonlinear approximation; bit level mapping;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

One of the difficulties encountered in realizing artificial neural network based on VLSI is the choice of the implementation method of activation function. At present, the main approaches to solve this problem are piecewise nonlinear approximation and bit level mapping. Based on hyperbolic tangent, the final output error of the two methods is discussed through the hardware implementation and software analysis of the artificial neural network nodes. We found that the nonlinear approximation method has the problem of large output fluctuation, and the amplification effect of the backpropagation can not be ignored. Therefore, this paper proposes that the bit level mapping method has more advantages in practical applications in the implementation of high-precision artificial neural nodes.

引用

页码：278 / 281

页数：4

共 50 条

[41] Increasing Accuracy of Power Consumption Using Artificial Neural Network
Syukur, Arry Muhammad
Putrada, Aji Gautama
Abdurohman, Maman
2019 7TH INTERNATIONAL CONFERENCE ON INFORMATION AND COMMUNICATION TECHNOLOGY (ICOICT), 2019, : 92 - 97
[42] Artificial neural network for violation analysis
Zhang, Z
Polet, P
Vanderhaegen, F
Millot, P
RELIABILITY ENGINEERING & SYSTEM SAFETY, 2004, 84 (01) : 3 - 18
[43] Simplified Hardware Implementation of Memoryless Dot Product for Neural Network Inference
Kouretas, I
Paliouras, V
2021 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS), 2021,
[44] Secured Node Identification Approach Based on Artificial Neural Network Infrastructure for Wireless Sensor Networks
Caleb, S.
Thangaraj, S. John Justin
2023 9TH INTERNATIONAL CONFERENCE ON ELECTRICAL ENERGY SYSTEMS, ICEES, 2023, : 646 - 651
[45] Influence and analysis of artificial neural network parameters on system identification accuracy and determination method of parameters
College of Automotive, Tongji University, Shanghai 201804, China
不详
不详
Jixie Gongcheng Xuebao, 2006, 7 (217-221+226): : 217 - 221+226
[46] Digital Artificial Neural Network Implementation on a FPGA for Data Classification
Morales, C.
Flores, U.
Adam, M.
Diaz, M.
Caballero, J. A.
Criado, D.
Pavoni, S.
IEEE LATIN AMERICA TRANSACTIONS, 2015, 13 (10) : 3216 - 3220
[47] Parallel implementation of Artificial Neural Network training for speech recognition
Scanzio, Stefano
Cumani, Sandro
Gemello, Roberto
Mana, Franco
Laface, P.
PATTERN RECOGNITION LETTERS, 2010, 31 (11) : 1302 - 1309
[48] Uncertainty analysis of sensitivity of MEMS microphone based on artificial neural network
Liu, Lei
Jia, Renxu
IEICE ELECTRONICS EXPRESS, 2019, 16 (24)
[49] Analysis and Prediction of Electrospun Nanofiber Diameter Based on Artificial Neural Network
Ma, Ming
Zhou, Huchen
Gao, Suhan
Li, Nan
Guo, Wenjuan
Dai, Zhao
POLYMERS, 2023, 15 (13)
[50] Detection of sleep breathing sound based on artificial neural network analysis
Emoto, Takahiro
Abeyratne, Udantha R.
Kawano, Kenichiro
Okada, Takuya
Jinnouchi, Osamu
Kawata, Ikuji
BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2018, 41 : 81 - 89

← 1 2 3 4 5 →