Ex-situ Training of Dense Memristor Crossbar for Neuromorphic Applications

被引：0
|
作者：
Hasan, Raqibul [1 ]
Yakopcic, Chris [1 ]
Taha, Tarek M. [1 ]
机构：
[1] Univ Dayton, Dept Elect & Comp Engn, Dayton, OH 45469 USA
来源：
PROCEEDINGS OF THE 2015 IEEE/ACM INTERNATIONAL SYMPOSIUM ON NANOSCALE ARCHITECTURES (NANOARCH 15) | 2015年
关键词：
Memristor crossbars; deep neural networks; pattern recognition; low power system;
D O I：
暂无
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
This study proposes a technique for programming a dense memristor crossbar array without isolation transistors (0T1M) in order to achieve ex-situ training of a neural network. Programming memristors to a specific resistance level requires an iterative process needing the reading of individual memristor resistances due to memristor device stochasticity. This paper presents a circuit to read individual resistances from a 0T1M crossbar and a method to map neuron synaptic weights into a novel neural circuit to enable ex-situ training. The results show that we are able to train the resistances in a 0T1M crossbar and that the 0T1M system is about 93% smaller in area than 1T1M systems.
引用
收藏
页码：75 / 81
页数：7
相关论文
共 50 条
  • [1] Ex-Situ Programming in a Neuromorphic Memristor Based Crossbar Circuit
    Yakopcic, Chris
    Taha, Tarek M.
    [J]. PROCEEDINGS OF THE 2015 IEEE NATIONAL AEROSPACE AND ELECTRONICS CONFERENCE (NAECON), 2015, : 300 - 304
  • [2] Ex-situ training of large memristor crossbars for neural network applications
    Hasan, Raqibul
    Yakopcic, Chris
    Taha, Tarek M.
    [J]. ANALOG INTEGRATED CIRCUITS AND SIGNAL PROCESSING, 2019, 99 (01) : 1 - 10
  • [3] Ex-situ training of large memristor crossbars for neural network applications
    Raqibul Hasan
    Chris Yakopcic
    Tarek M. Taha
    [J]. Analog Integrated Circuits and Signal Processing, 2019, 99 : 1 - 10
  • [4] Learning Method for Ex-situ Training of Memristor Crossbar based Multi-Layer Neural Network
    Bala, Anu
    Adeyemo, Adedotun
    Yang, Xiaohan
    Jabir, Abusaleh
    [J]. 2017 9TH INTERNATIONAL CONGRESS ON ULTRA MODERN TELECOMMUNICATIONS AND CONTROL SYSTEMS AND WORKSHOPS (ICUMT), 2017, : 305 - 310
  • [5] Memristor Based Neuromorphic Circuit for Ex-Situ Training of Multi-Layer Neural Network Algorithms
    Yakopcic, Chris
    Hasan, Raqibul
    Taha, Tarek M.
    [J]. 2015 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2015,
  • [6] Method for ex-situ training in memristor-based neuromorphic circuit using robust weight programming method
    Yakopcic, C.
    Taha, T. M.
    McLean, M.
    [J]. ELECTRONICS LETTERS, 2015, 51 (12) : 899 - 900
  • [7] Neuromorphic Computing with Memristor Crossbar
    Zhang, Xinjiang
    Huang, Anping
    Hu, Qi
    Xiao, Zhisong
    Chu, Paul K.
    [J]. PHYSICA STATUS SOLIDI A-APPLICATIONS AND MATERIALS SCIENCE, 2018, 215 (13):
  • [8] Enabling Back Propagation Training of Memristor Crossbar Neuromorphic Processors
    Hasan, Raqibul
    Taha, Tarek M.
    [J]. PROCEEDINGS OF THE 2014 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2014, : 21 - 28
  • [9] Impact of Memristor Switching Noise in a Neuromorphic Crossbar
    Yakopcic, Chris
    Taha, Tarek M.
    Subramanyam, Guru
    Pino, Robinson E.
    [J]. PROCEEDINGS OF THE 2015 IEEE NATIONAL AEROSPACE AND ELECTRONICS CONFERENCE (NAECON), 2015, : 320 - 326
  • [10] Sequential Memristor Crossbar for Neuromorphic Pattern Recognition
    Son Ngoc Truong
    Khoa Van Pham
    Yang, Wonsun
    Min, Kyeong-Sik
    [J]. IEEE TRANSACTIONS ON NANOTECHNOLOGY, 2016, 15 (06) : 922 - 930
12345