CoChat

Reservoir Computing Using Diffusive Memristors

Open access

The DOI matches its Crossref record; 37 of the 38 references it cites were found in a registry or an index.

Abstract

Reservoir computing (RC) is a framework that can extract features from a temporal input into a higher‐dimension feature space. The reservoir is followed by a readout layer that can analyze the extracted features to accomplish tasks such as inference and classification. RC systems inherently exhibit an advantage, since the training is only performed at the readout layer, and therefore they are able to compute complicated temporal data with a low training cost. Herein, a physical reservoir computing system using diffusive memristor‐based reservoir and drift memristor‐based readout layer is experimentally implemented. The rich nonlinear dynamic behavior exhibited by a diffusive memristor due to Ag migration and the robust in situ training of drift memristor arrays makes the combined system ideal for temporal pattern classification. It is then demonstrated experimentally that the RC system can successfully identify handwritten digits from the Modified National Institute of Standards and Technology (MNIST) dataset, achieving an accuracy of 83%.

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 · 10.1002/aisy.201900084

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Reservoir computing system based on diffusive memristor: a) Schematic of an RC system, showing the reservoir with internal dynamics and a readout function. The weight matrix connecting the reservoir state and the output needs to be trained. b) Equivalent schematic of a simplified system where the reservoir is populated with nodes with recurrent connections having a magnitude less than 1. c) The conductivity of the diffusive memristor is influenced by the periodic voltage stimulation that is provided on its top electrode ( þ ) while grounding the bottom electrode (  ). In the top panel, in two consecutive time slots, a voltage stimulus is provided. This results in a continuous Ag filament and high conductivity when the device state is analyzed in the fourth time slot. In the middle panel, in three consecutive time slots, a voltage stimulus is applied resulting in a much thicker filament and even higher conductivity when the device state is analyzed in the fourth time slot. In case of the bottom panel, a voltage stimulus is applied only in the first time slot. When the device state is evaluated in the fourth time slot, the filament has already broken down resulting in very low conductivity. This is because the device has enough time to relax back to its initial high resistance state, owing to its volatile nature. The diffusive memristor is analogous to a node with a recurrent connection having a weight less than 1. Hence, it is always decaying its state in every time frame unless a sufficiently large input is provided to counteract the effect of the feedback.
Figure 1 Reservoir computing system based on diffusive memristor: a) Schematic of an RC system, showing the reservoir with internal dynamics and a readout function. The weight matrix connecting the reservoir state and the output needs to be trained. b) Equivalent schematic of a simplified system where the reservoir is populated with nodes with recurrent connections having a magnitude less than 1. c) The conductivity of the diffusive memristor is influenced by the periodic voltage stimulation that is provided on its top electrode ( þ ) while grounding the bottom electrode ( ). In the top panel, in two consecutive time slots, a voltage stimulus is provided. This results in a continuous Ag filament and high conductivity when the device state is analyzed in the fourth time slot. In the middle panel, in three consecutive time slots, a voltage stimulus is applied resulting in a much thicker filament and even higher conductivity when the device state is analyzed in the fourth time slot. In case of the bottom panel, a voltage stimulus is applied only in the first time slot. When the device state is evaluated in the fourth time slot, the filament has already broken down resulting in very low conductivity. This is because the device has enough time to relax back to its initial high resistance state, owing to its volatile nature. The diffusive memristor is analogous to a node with a recurrent connection having a weight less than 1. Hence, it is always decaying its state in every time frame unless a sufficiently large input is provided to counteract the effect of the feedback.CC-BY · From the PDF
Figure 2 Waveform Design for the Reservoir: First set data can be found in Figure S1, Supporting Information. a) Schematic illustration of second set waveforms. 100 μ s, 1.25 V pulses signify 1, whereas 0 V pulse signifies 0. The 4-bit patterns are encoded in the form of four pulses. The first pulse in the pattern signifies the most significant bit (MSB), whereas the last pulse in the pattern signifies the least significant bit (LSB). As an example, in the topmost pulse pattern, the MSB is high or ' 1, ' whereas the LSB is low or ' 0. ' Pre-Read pulses are applied to ensure that the device is at its initial state before the application of a new pulse. The horizontal dashed red line indicates the end of the bit pattern. After the application of the pulse pattern with a 4-bit pattern encoded, the state of the memristor is read at two points. The first one is a fast read that is done around 5 μ s after the pulse pattern ends, and the second one is a slow read that is done around 300 μ s after the pulse pattern ends. The top panel shows the pattern. The bottom three panels show the current responses of the reservoir to the applied bit patterns. The applied bit patterns are (from top to bottom) 1110, 1001, and 0011. b) Schematic illustration of third set waveforms. 100 μ s, 1.25 V pulses signify 1, whereas 0 V pulse signifies 0. In this set of experiments, a 700 μ s, 0.8 V excitatory pulse is applied at the beginning of the pattern to ensure that the device is in an already low resistance state. The 4-bit patterns are encoded in the form of four pulses. The fi rst pulse in the pattern signifies the MSB or the most significant bit, whereas the last pulse in the pattern signifies the LSB or the least significant bit. As an example, in the topmost pulse pattern, the MSB is high or ' 1, ' whereas the LSB is low or ' 0. ' Pre-Read pulses are applied to ensure that the device is at its initial state before the application of a new pulse. The horizontal dashed red line indicates the end of the bit pattern. After the application of the pulse pattern with a 4-bit pattern encoded, the state of the memristor is read at two points. The first one is a fast read that is done right after the pulse pattern ends, and the second one is a slow read that is done around 300 μ s after the pulse pattern ends. The top panel shows the pattern. The bottom three panels show the current responses of the reservoir to the applied bit patterns. The applied bit patterns are (from top to bottom) 1110, 1001, and 0011. c) The distribution of current responses of a diffusive memristor corresponding to all possible 4-bit inputs including those illustrated in panel (a). One hundred measurements were experimentally conducted for each 4-bit input at the fast read. d) The distribution of current responses of a diffusive memristor corresponding to all possible 4-bit inputs for this set of data including those shown in panel (b). One hundred measurements were experimentally conducted for each 4-bit input at the fast read.
Figure 2 Waveform Design for the Reservoir: First set data can be found in Figure S1, Supporting Information. a) Schematic illustration of second set waveforms. 100 μ s, 1.25 V pulses signify 1, whereas 0 V pulse signifies 0. The 4-bit patterns are encoded in the form of four pulses. The first pulse in the pattern signifies the most significant bit (MSB), whereas the last pulse in the pattern signifies the least significant bit (LSB). As an example, in the topmost pulse pattern, the MSB is high or ' 1, ' whereas the LSB is low or ' 0. ' Pre-Read pulses are applied to ensure that the device is at its initial state before the application of a new pulse. The horizontal dashed red line indicates the end of the bit pattern. After the application of the pulse pattern with a 4-bit pattern encoded, the state of the memristor is read at two points. The first one is a fast read that is done around 5 μ s after the pulse pattern ends, and the second one is a slow read that is done around 300 μ s after the pulse pattern ends. The top panel shows the pattern. The bottom three panels show the current responses of the reservoir to the applied bit patterns. The applied bit patterns are (from top to bottom) 1110, 1001, and 0011. b) Schematic illustration of third set waveforms. 100 μ s, 1.25 V pulses signify 1, whereas 0 V pulse signifies 0. In this set of experiments, a 700 μ s, 0.8 V excitatory pulse is applied at the beginning of the pattern to ensure that the device is in an already low resistance state. The 4-bit patterns are encoded in the form of four pulses. The fi rst pulse in the pattern signifies the MSB or the most significant bit, whereas the last pulse in the pattern signifies the LSB or the least significant bit. As an example, in the topmost pulse pattern, the MSB is high or ' 1, ' whereas the LSB is low or ' 0. ' Pre-Read pulses are applied to ensure that the device is at its initial state before the application of a new pulse. The horizontal dashed red line indicates the end of the bit pattern. After the application of the pulse pattern with a 4-bit pattern encoded, the state of the memristor is read at two points. The first one is a fast read that is done right after the pulse pattern ends, and the second one is a slow read that is done around 300 μ s after the pulse pattern ends. The top panel shows the pattern. The bottom three panels show the current responses of the reservoir to the applied bit patterns. The applied bit patterns are (from top to bottom) 1110, 1001, and 0011. c) The distribution of current responses of a diffusive memristor corresponding to all possible 4-bit inputs including those illustrated in panel (a). One hundred measurements were experimentally conducted for each 4-bit input at the fast read. d) The distribution of current responses of a diffusive memristor corresponding to all possible 4-bit inputs for this set of data including those shown in panel (b). One hundred measurements were experimentally conducted for each 4-bit input at the fast read.CC-BY · From the PDF
Figure 3 Schematic of the diffusive memristor-based dynamic reservoir for classifying MNIST-based temporal sequences. a) The original 28 -28 image is binarized and cropped to size 22 -20, before being divided into five columns and sequentially joined with each other. The resultant 110 -4 input is then converted to 110 4-bit temporal voltage patterns as inputs to diffusive memristors. The current responses of the diffusive memristors are used to in situ train the 1T1R crossbar array. b) Optical micrograph of the 128 -64 1-transistor 1-memristor (1T1R) crossbar with the probe card landed. The color blocks illustrate the sub-array used for implementing the readout layer, consisting of 110 -20 devices using differential pairs. c) Scanning electron micrograph of a single 1T1R cell in (b). 1T1R cells of the same row share the common bottom bit line, while those of the same column share the common word line and memristor top electrode (TE) line. Scale bar: 20 μ m.
Figure 3 Schematic of the diffusive memristor-based dynamic reservoir for classifying MNIST-based temporal sequences. a) The original 28 -28 image is binarized and cropped to size 22 -20, before being divided into five columns and sequentially joined with each other. The resultant 110 -4 input is then converted to 110 4-bit temporal voltage patterns as inputs to diffusive memristors. The current responses of the diffusive memristors are used to in situ train the 1T1R crossbar array. b) Optical micrograph of the 128 -64 1-transistor 1-memristor (1T1R) crossbar with the probe card landed. The color blocks illustrate the sub-array used for implementing the readout layer, consisting of 110 -20 devices using differential pairs. c) Scanning electron micrograph of a single 1T1R cell in (b). 1T1R cells of the same row share the common bottom bit line, while those of the same column share the common word line and memristor top electrode (TE) line. Scale bar: 20 μ m.CC-BY · From the PDF
Figure 4 In situ training of diffusive memristor reservoir with 1T1R fully connected array for MNIST classification. a) The in-batch accuracy along the 2-epoch in situ training. The experimental curve tightly follows the simulation. b) The loss along the training. c) The inference result on classifying the MNIST test set. d) The distribution of the weights of the 1T1R fully connected layer prior to and after the training.
Figure 4 In situ training of diffusive memristor reservoir with 1T1R fully connected array for MNIST classification. a) The in-batch accuracy along the 2-epoch in situ training. The experimental curve tightly follows the simulation. b) The loss along the training. c) The inference result on classifying the MNIST test set. d) The distribution of the weights of the 1T1R fully connected layer prior to and after the training.CC-BY · From the PDF

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

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

Research works printed but not linked: 5

  1. Jaeger H., 2001, German National Research Center for Information Technology GMD Technical Report, 13
  2. J. J.Steil in2004 IEEE Int. Joint Conf. on Neural Networks (IEEE Cat. No.04CH37541) Vol.842 IEEE Budapest2004 pp.843–848.
  3. Jaeger H., 2002, Tutorial on Training Recurrent Neural Networks, Covering BPPT, RTRL, EKF and the “Echo State Network” Approach
  4. J.Lin Annadi S.Sonde C.Chen L.Stan K. V. L. V.Achari S.Ramanathan S.Guha in2016 IEEE Int. Electron Devices Meeting (IEDM) IEEE San Francisco2016 pp.34.35.31–34.35.34.
  5. 10.1038/s41928-019-0221-6 · DOI · Paper page

Cited by

See all 289

Metadata

Authors
Midya, R., Wang, Z., Asapu, S., Zhang, X., Rao, M., Song, W., Zhuo, Y., Upadhyay, N., Xia, Q., & Yang, J. J.
Affiliations
University of Massachusetts Amherst (US)
Corresponding author
  • Xia, Qiangfei
  • Yang, J. Joshua
Venue
Advanced Intelligent Systems
Publisher
Wiley
Published
2019-09-25
Volume
1
Issue
7
ISSN
2640-4567 (ISSN-L)
OpenAlex
W2971405544
Cited by
279
References
38
Citations by year
2026: 36 · 2025: 49 · 2024: 59 · 2023: 55 · 2022: 36 · 2021: 26 · 2020: 15 · 2019: 2
FWCI
9.01
Type
journal article
Language
en
Open access
Locations
Advanced Intelligent Systems (publishedVersion, cc-by)
Funders
Air Force Office of Scientific Research (FA9550-19-1-0213)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Reservoir Computing
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
  • Physical Sciences › Mathematics › Theoretical Computer Science › DNA and Biological Computing
Keywords
reservoir computing, diffusive memristors, physical reservoir computing, nonlinear dynamics, in situ training, handwritten digit recognition, MNIST

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

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.