Hardware Implementation of Tanh Exponential Activation Function using FPGA

被引：9
|
作者：
Bouguezzi, Safa [1 ,2 ]
Faiedh, Hassene [1 ,3 ]
Souani, Chokri [1 ,3 ]
机构：
[1] Univ Monastir, Microelect & Instrumentat Lab, Monastir, Tunisia
[2] Univ Monastir, Fac Sci Monastir, Monastir, Tunisia
[3] Univ Sousse, Higher Inst Appl Sci & Technol Sousse, Sousse, Tunisia
来源：
2021 18TH INTERNATIONAL MULTI-CONFERENCE ON SYSTEMS, SIGNALS & DEVICES (SSD) | 2021年
关键词：
FPGA; activation function; Tanh Exponential (TanhExp); Piecewise Linear approximation (PWL); Convolutional Neural Network (CNN); hardware resources;
D O I：
10.1109/SSD52085.2021.9429506
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：
The most active research area for Field Programmable Gate Arrays is the Convolution Neural Network (CNN), and the gist of any CNN is an activation function. Therefore, various non-linear activation functions aloe required for deeper CNNo In this paper, we aim to implement the Tanh Exponential (TanhExp) activation function on Artix-7 and Zynq-7000. To this end, we will use the piecewise linear approximation and the second-order polynomial approximation while using the IEEE754 2008 floating-point representation. We present an investigation of the required hardware resources. We also evaluate the efficiency of each method of approximation and its derivative.
引用
收藏
页码：1020 / 1025
页数：6
相关论文
共 50 条
  • [41] Hardware implementation of radial-basis neural networks with Gaussian activation functions on FPGA
    Shymkovych, Volodymyr
    Telenyk, Sergii
    Kravets, Petro
    NEURAL COMPUTING & APPLICATIONS, 2021, 33 (15): : 9467 - 9479
  • [42] Linear fragmentation Morris–Lecar realization using new exponential module instead of hyperbolic function in FPGA implementation
    Ghiasi A.
    Zahedi A.
    Haghiri S.
    Journal of Ambient Intelligence and Humanized Computing, 2023, 14 (04) : 4355 - 4370
  • [43] Hardware implementation of radial-basis neural networks with Gaussian activation functions on FPGA
    Volodymyr Shymkovych
    Sergii Telenyk
    Petro Kravets
    Neural Computing and Applications, 2021, 33 : 9467 - 9479
  • [44] Hardware Implementation of Hyperbolic Tangent Activation Function for Floating Point Formats
    Arvind, T. K. R.
    Brand, Marcel
    Heidorn, Christian
    Boppu, Srinivas
    Hannig, Frank
    Teich, Rirgen
    2020 24TH INTERNATIONAL SYMPOSIUM ON VLSI DESIGN AND TEST (VDAT), 2020,
  • [45] Hardware implementation of Variable Precision Multiplication on FPGA
    Anane, N.
    Bessalah, H.
    Issad, M.
    Messaoudi, K.
    Anane, M.
    DTIS: 2009 4TH IEEE INTERNATIONAL CONFERENCE ON DESIGN & TECHNOLOGY OF INTEGRATED SYSTEMS IN NANOSCALE ERA, PROCEEDINGS, 2009, : 77 - +
  • [46] A Hardware FPGA Implementation of Fault Attack Countermeasure
    Mestiri, Hassen
    Kahri, Fatma
    Bouallegue, Belgacem
    Machhout, Mohsen
    201415TH INTERNATIONAL CONFERENCE ON SCIENCES & TECHNIQUES OF AUTOMATIC CONTROL & COMPUTER ENGINEERING (STA'2014), 2014, : 178 - 183
  • [47] Hardware implementation of Multi-Rate input SoftMax activation function
    Wasef, Michael
    Rafla, Nader
    2021 IEEE INTERNATIONAL MIDWEST SYMPOSIUM ON CIRCUITS AND SYSTEMS (MWSCAS), 2021, : 783 - 786
  • [48] A sum of absolute differences implementation in FPGA hardware
    Wong, S
    Vassiliadis, S
    Cotofana, S
    PROCEEDINGS OF THE 28TH EUROMICRO CONFERENCE, 2002, : 183 - 188
  • [49] An FPGA hardware implementation of the Rijndael block cipher
    Dhoha, Chorfi
    Ben Othman, Slim
    Ben Saoud, Slim
    IEEE DTIS: 2006 International Conference on Design & Test of Integrated Systems in Nanoscale Technology, Proceedings, 2006, : 351 - 354
  • [50] Hardware Implementation of IP Packet Filtering in FPGA
    Cholakoska, Ana
    Efnusheva, Danijela
    Kalendar, Marija
    PROCEEDINGS OF THE 7TH INTERNATIONAL CONFERENCE ON APPLIED INNOVATIONS IN IT, VOL 7, ISSUE 1, 2019, 7 (01): : 23 - 29
12345