CoChat

Improving the Recognition Accuracy of Memristive Neural Networks via Homogenized Analog Type Conductance Quantization

Open access
Chen, Qilai, Han, Tingting, Tang, Minghua, Zhang, Zhang, Zheng, Xuejun, Liu, Gang · 2020 · Micromachines

The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index.

Abstract

Conductance quantization (QC) phenomena occurring in metal oxide based memristors demonstrate great potential for high-density data storage through multilevel switching, and analog synaptic weight update for effective training of the artificial neural networks. Continuous, linear and symmetrical modulation of the device conductance is a critical issue in QC behavior of memristors. In this contribution, we employ the scanning probe microscope (SPM) assisted electrode engineering strategy to control the ion migration process to construct single conductive filaments in Pt/HfOx/Pt devices. Upon deliberate tuning and evolution of the filament, 32 half integer quantized conductance states in the 16 G0 to 0.5 G0 range with enhanced distribution uniformity was achieved. Simulation results revealed that the numbers of the available QC states and fluctuation of the conductance at each state play an important role in determining the overall performance of the neural networks. The 32-state QC behavior of the hafnium oxide device shows improved recognition accuracy approaching 90% for handwritten digits, based on analog type operation of the multilayer perception (MLP) neural network.

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 · 10.3390/mi11040427

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Flow chart of device fabrication assisted by scanning probe microscope technique.
Figure 1 Flow chart of device fabrication assisted by scanning probe microscope technique.CC BY · PubMed Central
Figure 2 (a) Current–voltage characteristics of Pt/HfOx/Pt memristor device showing resistive switching with an ON/OFF ratio exceeding 103. Inset shows the structure of the device, with electrode protrusion extending into the hafnium oxide switching layer. (b) Histogram of the device resistances in the ON and OFF state. (c) Continuous regulation of the device current in the negatively biased reset processes. (d) Evolution of the device conductance as a function of the voltage pulse stressing numbers.
Figure 2 (a) Current–voltage characteristics of Pt/HfOx/Pt memristor device showing resistive switching with an ON/OFF ratio exceeding 103. Inset shows the structure of the device, with electrode protrusion extending into the hafnium oxide switching layer. (b) Histogram of the device resistances in the ON and OFF state. (c) Continuous regulation of the device current in the negatively biased reset processes. (d) Evolution of the device conductance as a function of the voltage pulse stressing numbers.CC BY · PubMed Central
Figure 3 Evolution of the device conductance as a function of the pulse modulating numbers for (a) the pre-treated Pt/HfOx/Pt memristor device sample A with microelectrode protrusion and (b) the untreated control device sample B with flat Pt/HfOx interface. Histogram and Gaussian curve fitting of the device conductance distributions at the 5th and 29th pulse modulation stages for sample A (c,e) and sample B (d,f), respectively.
Figure 3 Evolution of the device conductance as a function of the pulse modulating numbers for (a) the pre-treated Pt/HfOx/Pt memristor device sample A with microelectrode protrusion and (b) the untreated control device sample B with flat Pt/HfOx interface. Histogram and Gaussian curve fitting of the device conductance distributions at the 5th and 29th pulse modulation stages for sample A (c,e) and sample B (d,f), respectively.CC BY · PubMed Central
Figure 4 Mathematically simulated distribution of the device conductance in the 16 G0 to 0.5 G0 range with a 0.25 G0 step for (a) the pre-treated Pt/HfOx/Pt memristor device sample A with microelectrode protrusion and (b) the untreated control device sample B with flat Pt/HfOx interface, respectively.
Figure 4 Mathematically simulated distribution of the device conductance in the 16 G0 to 0.5 G0 range with a 0.25 G0 step for (a) the pre-treated Pt/HfOx/Pt memristor device sample A with microelectrode protrusion and (b) the untreated control device sample B with flat Pt/HfOx interface, respectively.CC BY · PubMed Central
Figure 5 (a) Schematic flowchart for the simulation of supervised learning with multilayer perception (MLP) neural network for handwritten digit recognition. (b) Architecture of the MLP network. Plots of the recognition accuracy as a function of the increasing training epochs (c) with the available QC states numbers of 4, 16, 32 and 64 in sample A and (d) with 32 QC states in sample A, B, and ideal devices without conductance fluctuation.
Figure 5 (a) Schematic flowchart for the simulation of supervised learning with multilayer perception (MLP) neural network for handwritten digit recognition. (b) Architecture of the MLP network. Plots of the recognition accuracy as a function of the increasing training epochs (c) with the available QC states numbers of 4, 16, 32 and 64 in sample A and (d) with 32 QC states in sample A, B, and ideal devices without conductance fluctuation.CC 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 44 references

Research works printed but not linked: 1

  1. Govoreanu, B., Kar, G.S., Chen, Y.Y., Paraschiv, V., Kubicek, S., Fantini, A., and Jossart, N. (2011, January 5–7). 10 × 10 nm2 Hf/HfOx crossbar resistive RAM with excellent performance, reliability and low-energy operation. Proceedings of the IEEE International Electron Devices Meeting, Washington, DC, USA.

Cited by

See all 12

Metadata

Authors
Chen, Q., Han, T., Tang, M., Zhang, Z., Zheng, X., & Liu, G.
Affiliations
  • School of Mechanical Engineering, Xiangtan University, Xiangtan 411105, China
  • Hefei University of Technology (CN)
  • Xiangtan University (CN)
  • Shanghai Jiao Tong University (CN)
Corresponding author
  • Zheng, Xuejun
  • Liu, Gang
Venue
Micromachines
Publisher
MDPI AG
Published
2020-04-18
Volume
11
Issue
4
Pages
427
ISSN
2072-666X (ISSN-L)
OpenAlex
W3016668785
Cited by
9
References
44
Citations by year
2024: 2 · 2023: 2 · 2022: 3 · 2021: 1 · 2020: 1
FWCI
0.60
Type
journal article
Language
en
Open access
Locations
  • Micromachines (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Natural Science Foundation of China (61722407, 11832016, 61674153, 61974090, U19A2053, 51775471, 61674049, 11835008)
  • the National Key R&D Program of China (2017YFB0405600,2017YFF0105000)
  • the Natural Science Foundation of Shanghai (19ZR1474500)
  • Hunan Provincial Innovation Foundation for Postgraduate (No. CX2018B373)
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
memristive neural networks, conductance quantization, conductive filaments, electrode design, scanning probe microscopy, multilevel switching, handwritten digit recognition, multilayer perceptron

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: Improving the Recognition Accuracy of Memristive Neural Networks via Homogenized Analog Type Conductance Quantization. https://app.cochat.ai/papers/10.3390/mi11040427
The paper
Chen, Q., Han, T., Tang, M., Zhang, Z., Zheng, X., & Liu, G. (2020). Improving the Recognition Accuracy of Memristive Neural Networks via Homogenized Analog Type Conductance Quantization. Micromachines. https://doi.org/10.3390/mi11040427

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.