skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Impact of Linearity and Write Noise of Analog Resistive Memory Devices in a Neural Algorithm Accelerator

Abstract

Resistive memory (ReRAM) shows promise for use as an analog synapse element in energy-efficient neural network algorithm accelerators. A particularly important application is the training of neural networks, as this is the most computationally-intensive procedure in using a neural algorithm. However, training a network with analog ReRAM synapses can significantly reduce the accuracy at the algorithm level. In order to assess this degradation, analog properties of ReRAM devices were measured and hand-written digit recognition accuracy was modeled for the training using backpropagation. Bipolar filamentary devices utilizing three material systems were measured and compared: one oxygen vacancy system, Ta-TaO x, and two conducting metallization systems, Cu-SiO 2, and Ag/chalcogenide. Analog properties and conductance ranges of the devices are optimized by measuring the response to varying voltage pulse characteristics. Key analog device properties which degrade the accuracy are update linearity and write noise. Write noise may improve as a function of device manufacturing maturity, but write nonlinearity appears relatively consistent among the different device material systems and is found to be the most significant factor affecting accuracy. As a result, this suggests that new materials and/or fundamentally different resistive switching mechanisms may be required to improve device linearity and achieve higher algorithmmore » training accuracy.« less

Authors:
 [1];  [1];  [1];  [1];  [1];  [1];  [1];  [1];  [1]
  1. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Publication Date:
Research Org.:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Sandia National Laboratories, Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1429781
Report Number(s):
SAND-2017-5497J
653576
Grant/Contract Number:  
AC04-94AL85000
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
IOP Nanotechnology
Additional Journal Information:
Journal Name: IOP Nanotechnology
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING

Citation Formats

Jacobs-Gedrim, Robin B., Agarwal, Sapan, Knisely, Kathrine E., Stevens, Jim E., van Heukelom, Michael S., Hughart, David R., Niroula, John, James, Conrad D., and Marinella, Matthew J.. Impact of Linearity and Write Noise of Analog Resistive Memory Devices in a Neural Algorithm Accelerator. United States: N. p., 2017. Web. doi:10.1109/ICRC.2017.8123657.
Jacobs-Gedrim, Robin B., Agarwal, Sapan, Knisely, Kathrine E., Stevens, Jim E., van Heukelom, Michael S., Hughart, David R., Niroula, John, James, Conrad D., & Marinella, Matthew J.. Impact of Linearity and Write Noise of Analog Resistive Memory Devices in a Neural Algorithm Accelerator. United States. doi:10.1109/ICRC.2017.8123657.
Jacobs-Gedrim, Robin B., Agarwal, Sapan, Knisely, Kathrine E., Stevens, Jim E., van Heukelom, Michael S., Hughart, David R., Niroula, John, James, Conrad D., and Marinella, Matthew J.. Fri . "Impact of Linearity and Write Noise of Analog Resistive Memory Devices in a Neural Algorithm Accelerator". United States. doi:10.1109/ICRC.2017.8123657.
@article{osti_1429781,
title = {Impact of Linearity and Write Noise of Analog Resistive Memory Devices in a Neural Algorithm Accelerator},
author = {Jacobs-Gedrim, Robin B. and Agarwal, Sapan and Knisely, Kathrine E. and Stevens, Jim E. and van Heukelom, Michael S. and Hughart, David R. and Niroula, John and James, Conrad D. and Marinella, Matthew J.},
abstractNote = {Resistive memory (ReRAM) shows promise for use as an analog synapse element in energy-efficient neural network algorithm accelerators. A particularly important application is the training of neural networks, as this is the most computationally-intensive procedure in using a neural algorithm. However, training a network with analog ReRAM synapses can significantly reduce the accuracy at the algorithm level. In order to assess this degradation, analog properties of ReRAM devices were measured and hand-written digit recognition accuracy was modeled for the training using backpropagation. Bipolar filamentary devices utilizing three material systems were measured and compared: one oxygen vacancy system, Ta-TaOx, and two conducting metallization systems, Cu-SiO2, and Ag/chalcogenide. Analog properties and conductance ranges of the devices are optimized by measuring the response to varying voltage pulse characteristics. Key analog device properties which degrade the accuracy are update linearity and write noise. Write noise may improve as a function of device manufacturing maturity, but write nonlinearity appears relatively consistent among the different device material systems and is found to be the most significant factor affecting accuracy. As a result, this suggests that new materials and/or fundamentally different resistive switching mechanisms may be required to improve device linearity and achieve higher algorithm training accuracy.},
doi = {10.1109/ICRC.2017.8123657},
journal = {IOP Nanotechnology},
number = ,
volume = ,
place = {United States},
year = {Fri Dec 01 00:00:00 EST 2017},
month = {Fri Dec 01 00:00:00 EST 2017}
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on December 1, 2018
Publisher's Version of Record

Save / Share: