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

The DOI matches its Crossref record; 52 of the 52 references it cites were found in a registry or an index.

Efficient and self-adaptive in-situ learning in multilayer memristor neural networks

Open access

Abstract

Memristors with tunable resistance states are emerging building blocks of artificial neural networks. However, in situ learning on a large-scale multiple-layer memristor network has yet to be demonstrated because of challenges in device property engineering and circuit integration. Here we monolithically integrate hafnium oxide-based memristors with a foundry-made transistor array into a multiple-layer neural network. We experimentally demonstrate in situ learning capability and achieve competitive classification accuracy on a standard machine learning dataset, which further confirms that the training algorithm allows the network to adapt to hardware imperfections. Our simulation using the experimental parameters suggests that a larger network would further increase the classification accuracy. The memristor neural network is a promising hardware platform for artificial intelligence with high speed-energy efficiency.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-018-04484-2

Cites

52 of 52 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Memristive platform for in situ learning. a An optical image of a wafer with transistor arrays. b Close-up of chip image showing arrays of various sizes. c Microscope image showing the 1T1R (one transistor one memristor) structure of the cell. Scale bar, 10 µm. d Cross-sectional scanning electron microscopic image of an individual 1T1R cell, which is cut in a focused ion beam microscope from the dashed line in c. Scale bar, 2 µm. e Cross-sectional transmission electron microscopic image of the integrated Ta/HfO2/Pt memristor. Scale bar, 2 nm. f All responsive devices over 20 potentiation/depression epochs of 200 pulses each. g Evolution of conductance during 20 cycles of full potentiation and depression for a single cell with 200 pulses per cycle, showing low cycle-to-cycle variability. More results are shown in Supplementary Fig. 1. h Evolution of conductance over one 200-pulse cycle of full potentiation and depression for all responsive devices in the array, with median conductance indicated by the yellow line. i Conductance of a 128 × 64 array after single-pulse conductance writing of the discrete Fourier transform matrix. Several stuck devices are visible (in yellow)
Fig. 1 Memristive platform for in situ learning. a An optical image of a wafer with transistor arrays. b Close-up of chip image showing arrays of various sizes. c Microscope image showing the 1T1R (one transistor one memristor) structure of the cell. Scale bar, 10 µm. d Cross-sectional scanning electron microscopic image of an individual 1T1R cell, which is cut in a focused ion beam microscope from the dashed line in c. Scale bar, 2 µm. e Cross-sectional transmission electron microscopic image of the integrated Ta/HfO2/Pt memristor. Scale bar, 2 nm. f All responsive devices over 20 potentiation/depression epochs of 200 pulses each. g Evolution of conductance during 20 cycles of full potentiation and depression for a single cell with 200 pulses per cycle, showing low cycle-to-cycle variability. More results are shown in Supplementary Fig. 1. h Evolution of conductance over one 200-pulse cycle of full potentiation and depression for all responsive devices in the array, with median conductance indicated by the yellow line. i Conductance of a 128 × 64 array after single-pulse conductance writing of the discrete Fourier transform matrix. Several stuck devices are visible (in yellow)CC BY · PubMed Central
Fig. 2 In situ training algorithm. a Schematic diagram of a two-layer neural network. Each neuron computes a weighted sum of its inputs and applies a nonlinear activation function. b The implementation of the network with a set of memristor crossbars. Each synaptic weight (arrows in a) corresponds to the conductance difference between two memristors (as illustrated by the orange columns). Each crossbar computes weighted sums of its input voltages. Between the crossbars is a layer of circuits that read the current from each wire, convert it to a voltage, and apply the activation function. The activation function was implemented in software in this work. c Flow chart of the in situ training. Steps in green boxes were implemented in hardware in this work, while those in yellow boxes were computationally expensive steps that can be accomplished with circuits integrated onto the chip in the future. The algorithm is described in detail in Methods
Fig. 2 In situ training algorithm. a Schematic diagram of a two-layer neural network. Each neuron computes a weighted sum of its inputs and applies a nonlinear activation function. b The implementation of the network with a set of memristor crossbars. Each synaptic weight (arrows in a) corresponds to the conductance difference between two memristors (as illustrated by the orange columns). Each crossbar computes weighted sums of its input voltages. Between the crossbars is a layer of circuits that read the current from each wire, convert it to a voltage, and apply the activation function. The activation function was implemented in software in this work. c Flow chart of the in situ training. Steps in green boxes were implemented in hardware in this work, while those in yellow boxes were computationally expensive steps that can be accomplished with circuits integrated onto the chip in the future. The algorithm is described in detail in MethodsCC BY · PubMed Central
Fig. 3 In situ online training and inference experiments on Modified National Institute of Standards and Technology (MNIST) handwritten digit recognition. a Typical handwritten digits from the MNIST database. b Photo of the integrated 128 × 64 array during measurement. The array was partitioned into two parts for the first and second layers, respectively. In all, 54 hidden neurons were used, so the first layer weight matrix is 64 × 54 (implemented using 6912 memristors) and the second layer matrix is 54 × 10 (implemented using 1080 memristors). The blue and green false-colored areas are the positive and negative parts of the differential pairs. c Minibatch accuracy increases over the course of training. Experimental data followed the defect-free simulation closely, with a consistent 2–4% gap. d The conductance-gate voltage relation extracted from data collected during training. The conductance was read using the scheme described in the Methods. The conductance includes the effects of sneak-paths and wire resistance, which makes the measured values smaller and the variance larger than those in Fig. 1b, c. The dashed line indicates the mean conductance, while the error bars show a 95% confidence interval for the measured conductance. The real-time online training accuracy with the readout weight values is shown in an animation in Supplementary Movie 1. e–g Typical correctly classified digit “9” and h–j misclassified digit “8” after the in situ training. e, h Images of the actual digits from the MNIST test set used as the input to the network. f, i The raw current measured from the output layer neurons. The neuron representing the digit “9” has the highest output current, indicating a correct classification. g, j The corresponding Bayesian probability of each digit, as calculated by a softmax function. More inference samples are shown in Supplementary Fig. 7 and Supplementary Movies 2 and 3
Fig. 3 In situ online training and inference experiments on Modified National Institute of Standards and Technology (MNIST) handwritten digit recognition. a Typical handwritten digits from the MNIST database. b Photo of the integrated 128 × 64 array during measurement. The array was partitioned into two parts for the first and second layers, respectively. In all, 54 hidden neurons were used, so the first layer weight matrix is 64 × 54 (implemented using 6912 memristors) and the second layer matrix is 54 × 10 (implemented using 1080 memristors). The blue and green false-colored areas are the positive and negative parts of the differential pairs. c Minibatch accuracy increases over the course of training. Experimental data followed the defect-free simulation closely, with a consistent 2–4% gap. d The conductance-gate voltage relation extracted from data collected during training. The conductance was read using the scheme described in the Methods. The conductance includes the effects of sneak-paths and wire resistance, which makes the measured values smaller and the variance larger than those in Fig. 1b, c. The dashed line indicates the mean conductance, while the error bars show a 95% confidence interval for the measured conductance. The real-time online training accuracy with the readout weight values is shown in an animation in Supplementary Movie 1. e–g Typical correctly classified digit “9” and h–j misclassified digit “8” after the in situ training. e, h Images of the actual digits from the MNIST test set used as the input to the network. f, i The raw current measured from the output layer neurons. The neuron representing the digit “9” has the highest output current, indicating a correct classification. g, j The corresponding Bayesian probability of each digit, as calculated by a softmax function. More inference samples are shown in Supplementary Fig. 7 and Supplementary Movies 2 and 3CC BY · PubMed Central
Fig. 4 Analysis based on experimental-calibrated simulation. a The experimental classification error (accuracy shown in Supplementary Fig. 10) matches the simulated accuracy. The simulation considers experiment parameters, including 11% devices stuck at 10 μS, 2% conductance update variation, limited conductance dynamic range, etc. The simulation on defect-free assumption shows an accuracy approaching that from TensorFlow. Each data point is the classification error rate on the complete testing set (10 000 images) after 500 images (simulation or TensorFlow) or 5000 images (experiment). b The impact of non-responsive devices on the inference accuracy with in situ and ex situ training approach. The non-responsive device was stuck in a very low-conductance state (10 µS), which is the typical defect device value observed in the experiment. The result shows that the in situ training process adapts to the defects, providing a much higher defect tolerance compared with pre-loading ex situ training weights into the network. With 50% stuck OFF devices, the network can still achieve over 60% accuracy. The error bar shows the s.d. over 10 simulations. c The multilayer network also helps with the defect tolerance. If one device is stuck, the associated hidden neuron will adjust the connections accordingly. The error bar shows the s.d. over 10 simulations. d The simulation of a larger network constructed on a larger memristor crossbar (1024 × 512) with experimental parameters (e.g., 11% defect rate) could achieve accuracy above 97%, which suggests a large memristor network could narrow the accuracy performance gap from the conventional CMOS hardware. The network architecture is shown in Supplementary Fig. 14
Fig. 4 Analysis based on experimental-calibrated simulation. a The experimental classification error (accuracy shown in Supplementary Fig. 10) matches the simulated accuracy. The simulation considers experiment parameters, including 11% devices stuck at 10 μS, 2% conductance update variation, limited conductance dynamic range, etc. The simulation on defect-free assumption shows an accuracy approaching that from TensorFlow. Each data point is the classification error rate on the complete testing set (10 000 images) after 500 images (simulation or TensorFlow) or 5000 images (experiment). b The impact of non-responsive devices on the inference accuracy with in situ and ex situ training approach. The non-responsive device was stuck in a very low-conductance state (10 µS), which is the typical defect device value observed in the experiment. The result shows that the in situ training process adapts to the defects, providing a much higher defect tolerance compared with pre-loading ex situ training weights into the network. With 50% stuck OFF devices, the network can still achieve over 60% accuracy. The error bar shows the s.d. over 10 simulations. c The multilayer network also helps with the defect tolerance. If one device is stuck, the associated hidden neuron will adjust the connections accordingly. The error bar shows the s.d. over 10 simulations. d The simulation of a larger network constructed on a larger memristor crossbar (1024 × 512) with experimental parameters (e.g., 11% defect rate) could achieve accuracy above 97%, which suggests a large memristor network could narrow the accuracy performance gap from the conventional CMOS hardware. The network architecture is shown in Supplementary Fig. 14CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 49 references

Research works printed but not linked: 2

  1. LeCun, Y. A theoretical framework of back-propagation. In Proc. of Connectionist Models Summer School (eds Touretzky, D., Hinton, G. E. & Sejnowski, T. J.) 21–28 (CMU, Pittsburgh, PA, USA, 1988).
  2. Gokmen, T. & Vlasov, Y. Acceleration of deep neural network training with resistive cross-point devices: design considerations. Front. Neurosci. 10, 333 (2016). · DOI · Paper page

Cited by

See all 931

Metadata

Authors
Li, C., Belkin, D., Li, Y., Yan, P., Hu, M., Ge, N., Jiang, H., Montgomery, E., Lin, P., Wang, Z., Song, W., Strachan, J. P., Barnell, M., Wu, Q., Williams, R. S., Yang, J. J., & Xia, Q.
Affiliations
  • University of Massachusetts Amherst (US)
  • Swarthmore College (US)
  • Binghamton University (US)
  • Hewlett-Packard (United States) (US)
  • U.S. Air Force Research Laboratory Information Directorate (US)
Corresponding author
  • Yang, J. Joshua
  • Xia, Qiangfei
Venue
Nature Communications
Published
2018-06-19
Volume
9
Issue
1
Pages
2385-2385
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2807750997
Cited by
931
References
49
Citations by year
2026: 76 · 2025: 112 · 2024: 103 · 2023: 131 · 2022: 120 · 2021: 126 · 2020: 147 · 2019: 100 · 2018: 16
FWCI
37.17
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Air Force Research Laboratory (FA8750-15-2-0044)
  • National Science Foundation (ECCS-1253073)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Ferroelectric and Negative Capacitance Devices
  • Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
Keywords
memristor neural networks, in situ learning, multilayer neural networks, monolithic integration, energy-efficient hardware, neuromorphic computing, classification accuracy

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.

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: Efficient and self-adaptive in-situ learning in multilayer memristor neural networks. https://app.cochat.ai/papers/10.1038/s41467-018-04484-2
The paper
Li, C., Belkin, D., Li, Y., Yan, P., Hu, M., Ge, N., Jiang, H., Montgomery, E., Lin, P., Wang, Z., Song, W., Strachan, J. P., Barnell, M., Wu, Q., Williams, R. S., Yang, J. J., & Xia, Q. (2018). Efficient and self-adaptive in-situ learning in multilayer memristor neural networks. Nature Communications. https://doi.org/10.1038/s41467-018-04484-2

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.