Convolutional Neural Networks Based on RRAM Devices for Image Recognition and Online Learning Tasks

Zhen Dong, Zheng Zhou, Zefan Li, Chen Liu, Peng Huang, Lifeng Liu, Xiaoyan Liu, Jinfeng KangIEEE Transactions on Electron Devices 2018 · DOI

Abstract

In this paper, we devise and optimize schemes for the resistive random-access memory (RRAM)-based hardware implementation of convolutional neural networks (CNNs). The key achievements are as follows: 1) a specific circuit for CNN and corresponding operation methods is presented; 2) quantization schemes for utilizing binary RRAM or RRAM with multilevel resistances as synapses are proposed, and simulations show that our CNN system percentage accuracy for the MNIST data set using multilevel RRAM and 97% accuracy using binary RRAM as synapses; 3) the influence of factors such as the number and size of kernels, as well as the device conductance variation, on the final recognition accuracy is discussed. Three ways to reduce the hardware cost are also analyzed; and 4) we implement the online learning function using the developed CNN system with a binary spike-time-dependent plasticity protocol and achieve up to 94% accuracy for the online learning tasks.
BibTeX
@article{dong2018convolutional,
  author = {Zhen Dong and Zheng Zhou and Zefan Li and Chen Liu and Peng Huang and Lifeng Liu and Xiaoyan Liu and Jinfeng Kang},
  title = {{Convolutional Neural Networks Based on RRAM Devices for Image Recognition and Online Learning Tasks}},
  journal = {IEEE Transactions on Electron Devices},
  year = {2018},
  doi = {10.1109/ted.2018.2882779},
}

Loading related content...

Figure