English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Bidirectional non-filamentary RRAM as an analog neuromorphic synapse, part II: Impact of Al/Mo/Pr0.7Ca0.3MnO3 device characteristics on neural network training accuracy

Fumarola, A., Sidler, S., Moon, K., Jang, J., Shelby, R. M., Narayanan, P., et al. (2017). Bidirectional non-filamentary RRAM as an analog neuromorphic synapse, part II: Impact of Al/Mo/Pr0.7Ca0.3MnO3 device characteristics on neural network training accuracy. IEEE journal of the Electron Devices Society, 6, 169-178. doi:10.1109/JEDS.2017.2782184.

Item is

Basic

show hide
Genre: Journal Article

Files

show Files
hide Files
:
Bidirectional_Non-Filamentary_RRAM_as_an_Analog_Neuromorphic_Synapse_Part_II_Impact_of_Al_Mo_Pr0.7Ca0.3MnO3_Device_Characteristics_on_Neural_Network_Training_Accuracy.pdf (Publisher version), 5MB
Name:
Bidirectional_Non-Filamentary_RRAM_as_an_Analog_Neuromorphic_Synapse_Part_II_Impact_of_Al_Mo_Pr0.7Ca0.3MnO3_Device_Characteristics_on_Neural_Network_Training_Accuracy.pdf
Description:
-
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
2017
Copyright Info:
IEEE

Locators

show
hide
Locator:
https://doi.org/10.1109/JEDS.2017.2782184 (Publisher version)
Description:
-

Creators

show
hide
 Creators:
Fumarola, Alessandro1, Author              
Sidler, Severin2, Author
Moon, Kibong2, Author
Jang, Junwoo2, Author
Shelby, Robert M.2, Author
Narayanan, Pritish2, Author
Leblebici, Yusuf2, Author
Hwang, Hyunsang2, Author
Burr, Geoffrey W.2, Author
Affiliations:
1Nano-Systems from Ions, Spins and Electrons, Max Planck Institute of Microstructure Physics, Max Planck Society, ou_3287476              
2External Organizations, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Neuromorphic computing embraces the “device history” offered by many analog non-volatile memory (NVM) devices to implement the small weight changes computed by a gradient-descent learning algorithm such as backpropagation. Deterministic and stochastic imperfections in the conductance response of real NVM devices can be encapsulated for modeling within a pair of “jump-tables.” Such jump-tables describe the full cumulative distribution function of conductance-change at each device conductance value, for both weight potentiation (SET) and depression (RESET). First, using several types of artificially constructed jump-tables, we revisit the relative importance of deviations from an ideal NVM with perfectly linear conductance response. Then, using jump-tables measured on improved non-filamentary resistive RAM devices based on Pr0.7Ca0.3MnO3 [see companion paper], we simulate the effects of their nonlinear conductance response on the training of a three-layer fully connected neural network. We find that, despite the relatively large conductance changes exhibited by any Pr0.7Ca0.3MnO3 device when either potentiating from its lowest conductance state or depressing from its highest conductance states, neural network training accuracies of >90% can be achieved. Highest accuracies are achieved by programming both conductances on each timestep (“fully bidirectional”), with the improved conductance on/off ratio of Al/Mo/PCMO resulting in marked improvements in training and test accuracy. Further accuracy improvements can be obtained by tuning the relative learning rate for potentiation (SET) by a factor of 1.66× with respect to depression (RESET), to offset the slight asymmetry between the average size of the associated SET and RESET conductance changes. Finally, we show that the bidirectional programming of Al/Mo/PCMO can be used to implement high-density neuromorphic systems with a single conductance per synapse, at only a slight degradation to accuracy.

Details

show
hide
Language(s):
 Dates: 2017-12-11
 Publication Status: Published online
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: BibTex Citekey: P13651
DOI: 10.1109/JEDS.2017.2782184
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: IEEE journal of the Electron Devices Society
  Other : IEEE journal of the Electron Devices Society : J-EDS
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: New York, NY : IEEE
Pages: - Volume / Issue: 6 Sequence Number: - Start / End Page: 169 - 178 Identifier: ISSN: 2168-6734
CoNE: https://pure.mpg.de/cone/journals/resource/2168-6734