Affordable Access

Access to the full text

Memristive GAN in Analog

Authors
  • Krestinskaya, O.1
  • Choubey, B.2
  • James, A. P.3
  • 1 Unaffiliated, Nur-Sultan, Kazakhstan , Nur-Sultan (Kazakhstan)
  • 2 Siegen University, Siegen, 57080, Germany , Siegen (Germany)
  • 3 Artificial General Intelligence and Neuromorphic Systems (NeuroAGI), Indian Institute of Information Technology and Management - Kerala, Trivandrum, Kerala, 695584, India , Trivandrum (India)
Type
Published Article
Journal
Scientific Reports
Publisher
Springer Nature
Publication Date
Apr 03, 2020
Volume
10
Issue
1
Identifiers
DOI: 10.1038/s41598-020-62676-7
Source
Springer Nature
License
Green

Abstract

Generative Adversarial Network (GAN) requires extensive computing resources making its implementation in edge devices with conventional microprocessor hardware a slow and difficult, if not impossible task. In this paper, we propose to accelerate these intensive neural computations using memristive neural networks in analog domain. The implementation of Analog Memristive Deep Convolutional GAN (AM-DCGAN) using Generator as deconvolutional and Discriminator as convolutional memristive neural network is presented. The system is simulated at circuit level with 1.7 million memristor devices taking into account memristor non-idealities, device and circuit parameters. The design is modular with crossbar arrays having a minimum average power consumption per neural computation of 47nW. The design exclusively uses the principles of neural network dropouts resulting in regularization and lowering the power consumption. The SPICE level simulation of GAN is performed with 0.18 μm CMOS technology and WOx memristive devices with RON = 40 kΩ and ROFF = 250 kΩ, threshold voltage 0.8 V and write voltage at 1.0 V.

Report this publication

Statistics

Seen <100 times