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

Title: Memristor design rules for dynamic learning and edge processing applications

Abstract

The ability to dynamically learn and adapt to changes in the environment is one of the hallmarks of biological systems. In this work, we identify the subset of the design space of memristive materials that is optimal for dynamic learning applications. Using an architecture inspired on the learning center of the insect brain, we implement a model system consisting of a discrete implementation of spiking neurons where dynamic learning takes place on a set of plastic synapses formed by memristor pairs in a crossbar array. Using two separate benchmarks, one comprising the dynamic learning of the Modified National Institute of Standards and Technology dataset and another one targeting one shot learning, we have identified the key properties that memristive materials should have to be optimal dynamic learners. The results obtained show that a fine degree of control of the memristor internal state is key to achieve high classification accuracy during dynamic learning but that within this optimal region learning is extremely robust both to device variability and to errors in the writing of the internal state, in all cases allowing for 2 sigma variations greater than 40% without significant loss of accuracy, hence overcoming two of the perceived limitations ofmore » memristors. By additionally requiring that learning takes place concurrently to information processing, we are able to derive a set constraints to the memristor dynamics.« less

Authors:
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
U.S. Department of Defense (DOD) - Defense Advanced Research Projects Agency (DARPA); USDOE Office of Science (SC)
OSTI Identifier:
1578165
DOE Contract Number:  
AC02-06CH11357
Resource Type:
Journal Article
Journal Name:
APL materials
Additional Journal Information:
Journal Volume: 7; Journal Issue: 9
Country of Publication:
United States
Language:
English

Citation Formats

Yanguas-Gil, A. Memristor design rules for dynamic learning and edge processing applications. United States: N. p., 2019. Web. doi:10.1063/1.5109910.
Yanguas-Gil, A. Memristor design rules for dynamic learning and edge processing applications. United States. doi:10.1063/1.5109910.
Yanguas-Gil, A. Sun . "Memristor design rules for dynamic learning and edge processing applications". United States. doi:10.1063/1.5109910.
@article{osti_1578165,
title = {Memristor design rules for dynamic learning and edge processing applications},
author = {Yanguas-Gil, A.},
abstractNote = {The ability to dynamically learn and adapt to changes in the environment is one of the hallmarks of biological systems. In this work, we identify the subset of the design space of memristive materials that is optimal for dynamic learning applications. Using an architecture inspired on the learning center of the insect brain, we implement a model system consisting of a discrete implementation of spiking neurons where dynamic learning takes place on a set of plastic synapses formed by memristor pairs in a crossbar array. Using two separate benchmarks, one comprising the dynamic learning of the Modified National Institute of Standards and Technology dataset and another one targeting one shot learning, we have identified the key properties that memristive materials should have to be optimal dynamic learners. The results obtained show that a fine degree of control of the memristor internal state is key to achieve high classification accuracy during dynamic learning but that within this optimal region learning is extremely robust both to device variability and to errors in the writing of the internal state, in all cases allowing for 2 sigma variations greater than 40% without significant loss of accuracy, hence overcoming two of the perceived limitations of memristors. By additionally requiring that learning takes place concurrently to information processing, we are able to derive a set constraints to the memristor dynamics.},
doi = {10.1063/1.5109910},
journal = {APL materials},
number = 9,
volume = 7,
place = {United States},
year = {2019},
month = {9}
}

Works referenced in this record:

Neuromorphic electronic systems
journal, January 1990


Pattern classification by memristive crossbar circuits using ex situ and in situ training
journal, June 2013

  • Alibart, Fabien; Zamanidoost, Elham; Strukov, Dmitri B.
  • Nature Communications, Vol. 4, Issue 1
  • DOI: 10.1038/ncomms3072

Neuromorphic computing using non-volatile memory
journal, October 2016


Biorealistic Implementation of Synaptic Functions with Oxide Memristors through Internal Ionic Dynamics
journal, June 2015


Integration of nanoscale memristor synapses in neuromorphic computing architectures
journal, September 2013


Training and operation of an integrated neuromorphic network based on metal-oxide memristors
journal, May 2015

  • Prezioso, M.; Merrikh-Bayat, F.; Hoskins, B. D.
  • Nature, Vol. 521, Issue 7550
  • DOI: 10.1038/nature14441

Nanoscale Memristor Device as Synapse in Neuromorphic Systems
journal, April 2010

  • Jo, Sung Hyun; Chang, Ting; Ebong, Idongesit
  • Nano Letters, Vol. 10, Issue 4, p. 1297-1301
  • DOI: 10.1021/nl904092h

The missing memristor found
journal, May 2008

  • Strukov, Dmitri B.; Snider, Gregory S.; Stewart, Duncan R.
  • Nature, Vol. 453, Issue 7191
  • DOI: 10.1038/nature06932

Memristive devices for computing
journal, January 2013

  • Yang, J. Joshua; Strukov, Dmitri B.; Stewart, Duncan R.
  • Nature Nanotechnology, Vol. 8, Issue 1, p. 13-24
  • DOI: 10.1038/nnano.2012.240

Memristive switching of single-component metallic nanowires
journal, March 2010


Bioinspired Programming of Memory Devices for Implementing an Inference Engine
journal, August 2015

  • Querlioz, Damien; Bichler, Olivier; Vincent, Adrien Francis
  • Proceedings of the IEEE, Vol. 103, Issue 8
  • DOI: 10.1109/jproc.2015.2437616

Memristor-Based Circuit Design for Multilayer Neural Networks
journal, February 2018

  • Zhang, Yang; Wang, Xiaoping; Friedman, Eby G.
  • IEEE Transactions on Circuits and Systems I: Regular Papers, Vol. 65, Issue 2
  • DOI: 10.1109/tcsi.2017.2729787

Learning in Memristor Crossbar-Based Spiking Neural Networks Through Modulation of Weight-Dependent Spike-Timing-Dependent Plasticity
journal, May 2018


A million spiking-neuron integrated circuit with a scalable communication network and interface
journal, August 2014


Loihi: A Neuromorphic Manycore Processor with On-Chip Learning
journal, January 2018


Circuit Elements With Memory: Memristors, Memcapacitors, and Meminductors
journal, October 2009

  • Di Ventra, Massimiliano; Pershin, Yuriy V.; Chua, Leon O.
  • Proceedings of the IEEE, Vol. 97, Issue 10
  • DOI: 10.1109/jproc.2009.2021077

The neuronal architecture of the mushroom body provides a logic for associative learning
journal, December 2014


Generating sparse and selective third-order responses in the olfactory system of the fly
journal, May 2010

  • Luo, S. X.; Axel, R.; Abbott, L. F.
  • Proceedings of the National Academy of Sciences, Vol. 107, Issue 23
  • DOI: 10.1073/pnas.1005635107

Mushroom body output neurons encode valence and guide memory-based action selection in Drosophila
journal, December 2014


Heterosynaptic Plasticity Underlies Aversive Olfactory Learning in Drosophila
journal, December 2015


Sparse coding with memristor networks
journal, May 2017

  • Sheridan, Patrick M.; Cai, Fuxi; Du, Chao
  • Nature Nanotechnology, Vol. 12, Issue 8
  • DOI: 10.1038/nnano.2017.83

Gradient-based learning applied to document recognition
journal, January 1998

  • Lecun, Y.; Bottou, L.; Bengio, Y.
  • Proceedings of the IEEE, Vol. 86, Issue 11
  • DOI: 10.1109/5.726791

A ferroelectric memristor
journal, September 2012

  • Chanthbouala, André; Garcia, Vincent; Cherifi, Ryan O.
  • Nature Materials, Vol. 11, Issue 10
  • DOI: 10.1038/nmat3415