Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 66 of the 67 references it cites were found in a registry or an index; 7 of the 30 sources its claims cite were checked for support.

Neuromorphic computing with multi-memristive synapses

Open access

Abstract

Neuromorphic computing has emerged as a promising avenue towards building the next generation of intelligent computing systems. It has been proposed that memristive devices, which exhibit history-dependent conductivity modulation, could efficiently represent the synaptic weights in artificial neural networks. However, precise modulation of the device conductance over a wide dynamic range, necessary to maintain high network accuracy, is proving to be challenging. To address this, we present a multi-memristive synaptic architecture with an efficient global counter-based arbitration scheme. We focus on phase change memory devices, develop a comprehensive model and demonstrate via simulations the effectiveness of the concept for both spiking and non-spiking neural networks. Moreover, we present experimental results involving over a million phase change memory devices for unsupervised learning of temporal correlations using a spiking neural network. The work presents a significant step towards the realization of large-scale and energy-efficient neuromorphic computing systems.

Overview

CoChat Summary

TL;DR

The paper proposes a multi-memristive synapse architecture that uses multiple memristive devices per synapse plus a counter-based arbitration scheme to improve dynamic range, resolution, and robustness for neuromorphic learning. It demonstrates the idea with PCM devices in simulations and in experiments on more than one million devices, including unsupervised temporal-correlation detection.

Key findings

  • In MNIST simulations for an ANN, the ideal floating-point classifier achieved 97.8% accuracy, while multi-memristive synapses reached >88.9% in the differential architecture and >90% in the non-differential architecture; the conventional 2-device differential architecture stayed below 15%.
  • In MNIST simulations for an SNN with unsupervised learning, the floating-point baseline achieved 77.2% accuracy; the network with multi-memristive synapses achieved >70% correct classification for N > 9, while the conventional 2-device differential architecture was below 21%.
  • In the small-scale temporal-correlation experiment, the number of misclassified inputs was 49 for N = 1, 8 for N = 3, and 0 for N = 7.
  • In the large-scale temporal-correlation experiment, 144,000 synapses with 7 PCM devices per synapse used 1,008,000 PCM devices total, and the experiment and simulation each reported about 0.1% misclassified inputs after training.

Methods

  • The paper combines device characterization, analytical modeling, simulations, and hardware experiments using a prototype PCM chip with 3 million devices in a four-bank inter-leaved architecture fabricated in 90 nm CMOS.
  • For simulation, the authors trained an ANN on 60,000 MNIST training images and tested on 10,000 MNIST test images; they also simulated an SNN for MNIST and an SNN for temporal-correlation detection using STDP-based learning rules.
  • The hardware correlation-detection experiment trained an SNN on 1000 synapses in the small-scale case and 144,000 synapses in the large-scale case, with some synapses receiving correlated inputs at correlation coefficient 0.75.

Limitations

  • The paper states that PCM devices have limiting characteristics including limited conductance range, asymmetric conductance response, nonlinear conductance response, stochasticity, and device-to-device variability.
  • The text makes evident that the scheme still relies on counter-based arbitration and may produce weight-dependent plasticity because device selection does not use exact conductance knowledge; the paper notes this could negatively affect ANN accuracy, though it may stabilize some SNNs.
  • The paper also states that its experimental data are available upon request only, which may limit immediate reproducibility from the text alone.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41467-018-04933-y

Cites

66 of 66 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 The multi-memristive synapse concept. a The net synaptic weight of a multi-memristive synapse is represented by the combined conductance (∑Gn) of multiple memristive devices. To realize synaptic efficacy, a read voltage signal, V, is applied to all devices. The resulting current flowing through each device is summed up to generate the synaptic output. b To capture synaptic plasticity, only one of the devices is selected at any instance of synaptic update. The synaptic update is induced by altering the conductance of the selected device as dictated by a learning algorithm. This is achieved by applying a suitable programming pulse to the selected device. c A counter-based arbitration scheme is used to select the devices that get programmed to achieve synaptic plasticity. A global selection counter whose maximum value is equal to the number of devices representing a synapse is used. At any instance of synaptic update, the device pointed to by the selection counter is programmed. Subsequently, the selection counter is incremented by a fixed amount. In addition to the selection counter, independent potentiation and depression counters can serve to control the frequency of the potentiation or depression events
Fig. 1 The multi-memristive synapse concept. a The net synaptic weight of a multi-memristive synapse is represented by the combined conductance (∑Gn) of multiple memristive devices. To realize synaptic efficacy, a read voltage signal, V, is applied to all devices. The resulting current flowing through each device is summed up to generate the synaptic output. b To capture synaptic plasticity, only one of the devices is selected at any instance of synaptic update. The synaptic update is induced by altering the conductance of the selected device as dictated by a learning algorithm. This is achieved by applying a suitable programming pulse to the selected device. c A counter-based arbitration scheme is used to select the devices that get programmed to achieve synaptic plasticity. A global selection counter whose maximum value is equal to the number of devices representing a synapse is used. At any instance of synaptic update, the device pointed to by the selection counter is programmed. Subsequently, the selection counter is incremented by a fixed amount. In addition to the selection counter, independent potentiation and depression counters can serve to control the frequency of the potentiation or depression eventsCC BY · PubMed Central
Fig. 2 Synapses based on phase change memory. a A PCM device consists of a phase-change material layer sandwiched between top and bottom electrodes. The crystalline region can gradually be increased by the application of potentiation pulses. A depression pulse creates an amorphous region that results in an abrupt drop in conductance, irrespective of the original state of the device. b Evolution of mean conductance as a function of the number of pulses for different programming current amplitudes (Iprog). Each curve is obtained by averaging the conductance measurements from 9700 devices. The inset shows a transmission electron micrograph of a characteristic PCM device used in this study. c Mean cumulative conductance change observed upon the application of repeated potentiation and depression pulses. The initial conductance of the devices is ∼5 μS. d The mean and the standard deviation (1σ) of the conductance values as a function of number of pulses for Iprog = 100 μA measured for 9700 devices and the corresponding model response for the same number of devices. The distribution of conductance after the 20th potentiation pulse and the corresponding distribution obtained with the model are shown in the inset. e The left panel shows a representative distribution of the conductance change induced by a single pulse applied at the same PCM device 1000 times. The pulse is applied as the 4th potentiation pulse to the device. The same measurement was repeated on 1000 different PCM devices, and the mean (μ) and standard deviation (σ) averaged over the 1000 devices are shown in the inset. The right panel shows a representative distribution of one conductance change induced by a single pulse on 1000 devices. The pulse is applied as the 4th potentiation pulse to the devices. The same measurement was repeated for 1000 conductance changes, and the mean and standard deviation averaged over the 1000 conductance changes are shown in the inset. It can be seen that the inter-device and the intra-device variability are comparable. The negative conductance changes are attributed to drift variability (see Supplementary Note 4)
Fig. 2 Synapses based on phase change memory. a A PCM device consists of a phase-change material layer sandwiched between top and bottom electrodes. The crystalline region can gradually be increased by the application of potentiation pulses. A depression pulse creates an amorphous region that results in an abrupt drop in conductance, irrespective of the original state of the device. b Evolution of mean conductance as a function of the number of pulses for different programming current amplitudes (Iprog). Each curve is obtained by averaging the conductance measurements from 9700 devices. The inset shows a transmission electron micrograph of a characteristic PCM device used in this study. c Mean cumulative conductance change observed upon the application of repeated potentiation and depression pulses. The initial conductance of the devices is ∼5 μS. d The mean and the standard deviation (1σ) of the conductance values as a function of number of pulses for Iprog = 100 μA measured for 9700 devices and the corresponding model response for the same number of devices. The distribution of conductance after the 20th potentiation pulse and the corresponding distribution obtained with the model are shown in the inset. e The left panel shows a representative distribution of the conductance change induced by a single pulse applied at the same PCM device 1000 times. The pulse is applied as the 4th potentiation pulse to the device. The same measurement was repeated on 1000 different PCM devices, and the mean (μ) and standard deviation (σ) averaged over the 1000 devices are shown in the inset. The right panel shows a representative distribution of one conductance change induced by a single pulse on 1000 devices. The pulse is applied as the 4th potentiation pulse to the devices. The same measurement was repeated for 1000 conductance changes, and the mean and standard deviation averaged over the 1000 conductance changes are shown in the inset. It can be seen that the inter-device and the intra-device variability are comparable. The negative conductance changes are attributed to drift variability (see Supplementary Note 4)CC BY · PubMed Central
Fig. 3 Multi-memristive synapses based on phase change memory. a The mean cumulative conductance change is experimentally obtained for synapses comprising 1, 3, and 7 PCM devices. The measurements are based on 1000 synapses, whereby each individual device is initialized to a conductance of ∼5 μS. For potentiation, a programming pulse of Iprog = 100 μA was used, whereas for depression, a programming pulse of Iprog = 450 μA was used. For depression, the conductance response can be made more symmetric by adjusting the length of the depression counter. b Distribution of the cumulative conductance change after the application of 10, 30, and 70 potentiation pulses to 1, 3, and 7 PCM synapses, respectively. The mean (μ) and the variance (σ2) scale almost linearly with the number of devices per synapse, leading to an improved weight update resolution
Fig. 3 Multi-memristive synapses based on phase change memory. a The mean cumulative conductance change is experimentally obtained for synapses comprising 1, 3, and 7 PCM devices. The measurements are based on 1000 synapses, whereby each individual device is initialized to a conductance of ∼5 μS. For potentiation, a programming pulse of Iprog = 100 μA was used, whereas for depression, a programming pulse of Iprog = 450 μA was used. For depression, the conductance response can be made more symmetric by adjusting the length of the depression counter. b Distribution of the cumulative conductance change after the application of 10, 30, and 70 potentiation pulses to 1, 3, and 7 PCM synapses, respectively. The mean (μ) and the variance (σ2) scale almost linearly with the number of devices per synapse, leading to an improved weight update resolutionCC BY · PubMed Central
Fig. 4 Applications of multi-memristive synapses in neural networks. a An artificial neural network is trained using backpropagation to perform handwritten digit classification. Bias neurons are used for the input and hidden neuron layers (white). A multi-memristive synapse model based on the nonlinear conductance response of PCM devices is used to represent the synaptic weights in these simulations. Increasing the number of devices in multi-memristive synapses (both in the differential and the non-differential architecture) improves the test accuracy. Simulations are repeated for five different weight initializations. The error bars represent the standard deviation (1σ). The dotted line shows the test accuracy obtained from a double-precision floating-point software implementation. b A spiking neural network is trained using an STDP-based learning rule for handwritten digit classification. Here again, a multi-memristive synapse model is used to represent the synaptic weights in simulations where the devices are arranged in the differential or the non-differential architecture. The classification accuracy of the network increases with the number of devices per synapse. Simulations are repeated for five different weight initializations. The error bars represent the standard deviation (1σ). The dotted line shows the test accuracy obtained from a double-precision floating-point implementation
Fig. 4 Applications of multi-memristive synapses in neural networks. a An artificial neural network is trained using backpropagation to perform handwritten digit classification. Bias neurons are used for the input and hidden neuron layers (white). A multi-memristive synapse model based on the nonlinear conductance response of PCM devices is used to represent the synaptic weights in these simulations. Increasing the number of devices in multi-memristive synapses (both in the differential and the non-differential architecture) improves the test accuracy. Simulations are repeated for five different weight initializations. The error bars represent the standard deviation (1σ). The dotted line shows the test accuracy obtained from a double-precision floating-point software implementation. b A spiking neural network is trained using an STDP-based learning rule for handwritten digit classification. Here again, a multi-memristive synapse model is used to represent the synaptic weights in simulations where the devices are arranged in the differential or the non-differential architecture. The classification accuracy of the network increases with the number of devices per synapse. Simulations are repeated for five different weight initializations. The error bars represent the standard deviation (1σ). The dotted line shows the test accuracy obtained from a double-precision floating-point implementationCC BY · PubMed Central
Fig. 5 Experimental demonstration of multi-memristive synapses used in a spiking neural network. a A spiking neural network is trained to perform the task of temporal correlation detection through unsupervised learning. Our network consists of 1000 multi-PCM synapses (in hardware) connected to one integrate-and-fire (I&F) software neuron. The synapses receive event-based data streams generated with Poisson distributions as presynaptic input spikes. 100 of the synapses receive correlated data streams with a correlation coefficient of 0.75, whereas the rest of the synapses receive uncorrelated data streams. The correlated and the uncorrelated data streams both have the same rate. The resulting postsynaptic outputs are accumulated at the neuronal membrane. The neuron fires, i.e., sends an output spike, if the membrane potential exceeds a threshold. The weight update amount is calculated using an exponential STDP rule based on the timing of the input spikes and the neuronal spikes. A potentiation (depression) pulse with fixed amplitude is applied if the desired weight change is higher (lower) than a threshold. b The synaptic weights are shown for synapses comprising N = 1, 3, and 7 PCM devices at the end of the experiment (5000 time steps). It can be seen that the weights of the synapses receiving correlated inputs tend to be larger than the weights of those receiving uncorrelated inputs. The weight distribution shows a clearer separation with increasing N. c Weight evolution of six synapses in the first 300 time steps of the experiment. The weight evolves more gradually with the number of devices per synapse. d Synaptic weight distribution of an SNN comprising 144,000 multi-PCM synapses with N = 7 PCM devices at the end of an experiment (3000 time steps) (upper panel). 14,400 synapses receive correlated input data streams with a correlation coefficient of 0.75. A total of 1,008,000 PCM devices are used for this large-scale experiment. The lower panel shows the synaptic weight distribution predicted by the PCM device model
Fig. 5 Experimental demonstration of multi-memristive synapses used in a spiking neural network. a A spiking neural network is trained to perform the task of temporal correlation detection through unsupervised learning. Our network consists of 1000 multi-PCM synapses (in hardware) connected to one integrate-and-fire (I&F) software neuron. The synapses receive event-based data streams generated with Poisson distributions as presynaptic input spikes. 100 of the synapses receive correlated data streams with a correlation coefficient of 0.75, whereas the rest of the synapses receive uncorrelated data streams. The correlated and the uncorrelated data streams both have the same rate. The resulting postsynaptic outputs are accumulated at the neuronal membrane. The neuron fires, i.e., sends an output spike, if the membrane potential exceeds a threshold. The weight update amount is calculated using an exponential STDP rule based on the timing of the input spikes and the neuronal spikes. A potentiation (depression) pulse with fixed amplitude is applied if the desired weight change is higher (lower) than a threshold. b The synaptic weights are shown for synapses comprising N = 1, 3, and 7 PCM devices at the end of the experiment (5000 time steps). It can be seen that the weights of the synapses receiving correlated inputs tend to be larger than the weights of those receiving uncorrelated inputs. The weight distribution shows a clearer separation with increasing N. c Weight evolution of six synapses in the first 300 time steps of the experiment. The weight evolves more gradually with the number of devices per synapse. d Synaptic weight distribution of an SNN comprising 144,000 multi-PCM synapses with N = 7 PCM devices at the end of an experiment (3000 time steps) (upper panel). 14,400 synapses receive correlated input data streams with a correlation coefficient of 0.75. A total of 1,008,000 PCM devices are used for this large-scale experiment. The lower panel shows the synaptic weight distribution predicted by the PCM device modelCC BY · PubMed Central

Verification

Claims the paper makes

5 Valid · 1 Not valid · 16 Not checked · 68 Own results and interpretations

55 more claims

Create a free account to see all 90 claims and to verify the 3 references still open.

See all claims — free account
  • It may state two separate claims as one. 18
  • Borderline: one of the tests for a substantial claim is unclear. 1
  • The quote only partly states this claim. 1
  • Probably not a substantial claim. 13

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 67 references

Research works printed but not linked: 1

  1. Burr, G. W. et al. Integration of nanoscale memristor synapses in neuromorphic computing architectures. Adv. Phys. X 2, 89–124 (2016).

Cited by

See all 899

Metadata

Authors
Boybat, I., Le Gallo, M., Nandakumar, S. R., Moraitis, T., Parnell, T., Tuma, T., Rajendran, B., Leblebici, Y., Sebastian, A., & Eleftheriou, E.
Affiliations
  • IBM Research - Zurich (CH)
  • New Jersey Institute of Technology (US)
  • École Polytechnique Fédérale de Lausanne (CH)
Corresponding author
  • Boybat, Irem
  • Sebastian, Abu
Venue
Nature Communications
Published
2018-06-28
Volume
9
Issue
1
Pages
2514-2514
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2769049661
Cited by
899
References
67
Citations by year
2026: 95 · 2025: 107 · 2024: 99 · 2023: 128 · 2022: 127 · 2021: 140 · 2020: 122 · 2019: 63 · 2018: 18
FWCI
32.77
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • Research Portal (King's College London) (submittedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Infoscience (Ecole Polytechnique Fédérale de Lausanne) (submittedVersion, cc-by-nc-nd)
  • arXiv (Cornell University)
Funders
  • National Science Foundation
  • European Commission (HORIZON2020, 682675)
  • Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
  • Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Reservoir Computing
  • Physical Sciences › Mathematics › Theoretical Computer Science › DNA and Biological Computing
Keywords
neuromorphic computing, memristors, phase-change memory, spiking neural networks, unsupervised learning, energy-efficient hardware, artificial neural networks
MeSH terms
Unsupervised Machine Learning, Action Potentials, Animals, Electric Conductivity, Electronics, Humans, Models, Neurological, Synapses, Neural Networks, Computer, Biomimetic Materials

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Neuromorphic computing with multi-memristive synapses. https://app.cochat.ai/papers/10.1038/s41467-018-04933-y
The paper
Boybat, I., Le Gallo, M., Nandakumar, S. R., Moraitis, T., Parnell, T., Tuma, T., Rajendran, B., Leblebici, Y., Sebastian, A., & Eleftheriou, E. (2018). Neuromorphic computing with multi-memristive synapses. Nature Communications. https://doi.org/10.1038/s41467-018-04933-y

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.