Implementation of Convolutional Neural Networks in Memristor Crossbar Arrays with Binary Activation and Weight Quantization
The DOI matches its Crossref record; 64 of the 64 references it cites were found in a registry or an index.
Abstract
We propose a hardware-friendly architecture of a convolutional neural network using a 32 × 32 memristor crossbar array having an overshoot suppression layer. The gradual switching characteristics in both set and reset operations enable the implementation of a 3-bit multilevel operation in a whole array that can be utilized as 16 kernels. Moreover, a binary activation function mapped to the read voltage and ground is introduced to evaluate the result of training with a boundary of 0.5 and its estimated gradient. Additionally, we adopt a fixed kernel method, where inputs are sequentially applied to a crossbar array with a differential memristor pair scheme, reducing unused cell waste. The binary activation has robust characteristics against device state variations, and a neuron circuit is experimentally demonstrated on a customized breadboard. Thanks to the analogue switching characteristics of the memristor device, the accurate vector-matrix multiplication (VMM) operations can be experimentally demonstrated by combining sequential inputs and the weights obtained through tuning operations in the crossbar array. In addition, the feature images extracted by VMM during the hardware inference operations on 100 test samples are classified, and the classification performance by off-chip training is compared with the software results. Finally, inference results depending on the tolerance are statistically verified through several tuning cycles.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“the memristive dot product engine (MDPE), which performs large-scale vector-matrix multiplication (VMM) operations using a memristive crossbar array6–13”
From Flexible self-rectifying synapse array for energy-efficient edge multiplication in electrocardiogram diagnosis (2025) · cites this paper as [11]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
Rastegari, M., Ordonez, V., Redmon, J., & Farhadi, A. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_32 · Cited by 4,408
Convolutional Neural Networks for Speech Recognition
Abdel-Hamid, O., Mohamed, A.-R., Jiang, H., Deng, L., Penn, G., & Yu, D. (2014) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2014.2339736 · Cited by 2,262
Fully hardware-implemented memristor convolutional neural network
Yao, P., Wu, H., Gao, B., Tang, J., Zhang, Q., Zhang, W., Yang, J. J., & Qian, H. (2020) · Nature · doi:10.1038/s41586-020-1942-4 · Cited by 2,179
Memristive crossbar arrays for brain-inspired computing
Xia, Q., & Yang, J. J. (2019) · Nature Materials · doi:10.1038/s41563-019-0291-x · Cited by 1,800
Analogue signal and image processing with large memristor crossbars
Li, C., Hu, M., Li, Y., Jiang, H., Ge, N., Montgomery, E., Zhang, J., Song, W., Dávila, N., Graves, C. E., et al. (2017) · Nature Electronics · doi:10.1038/s41928-017-0002-z · Cited by 1,292
Going Deeper in Spiking Neural Networks: VGG and Residual Architectures
Sengupta, A., Ye, Y., Wang, R., Liu, C., & Roy, K. (2019) · Frontiers in Neuroscience · doi:10.3389/fnins.2019.00095 · Cited by 1,211
Quantizing deep convolutional networks for efficient inference: A whitepaper
Krishnamoorthi, R. (2018) · arXiv · doi:10.48550/arxiv.1806.08342 · Cited by 753
Memristor‐Based Analog Computation and Neural Network Classification with a Dot Product Engine
Hu, M., Graves, C. E., Li, C., Li, Y., Ge, N., Montgomery, E., Davila, N., Jiang, H., Williams, R. S., Yang, J. J., et al. (2018) · Advanced Materials · doi:10.1002/adma.201705914 · Cited by 751
Research works printed but not linked: 4
- Sheu, S.S.; Chiang, P.C.; Lin, W.P.; Lee, H.Y.; Chen, P.S.; Chen, Y.S.; Wu, T.Y.; Chen, F. T.; Su, K.L.; Kao, M.J. A 5ns Fast Write Multi-level Non-volatile 1 K Bits RRAM Memory with Advance Write Scheme. In 2009 Symposium on VLSI Circuits ; IEEE, 2009; pp 82–83.
- Courbariaux, M.; Hubara, I.; Soudry, D.; El-Yaniv, R.; Bengio, Y. Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or–1 arXiv preprint arXiv:1602.030 2016.
- Courbariaux, M.; Bengio, Y.; David, J.P. Binaryconnect: Training deep neural networks with binary weights during propagations Adv. Neural Inf. Process. Syst. 2015; Vol. 28.
- Hubara I., 2017, J. Mach. Learn. Res., 18, 1
Cited by
Fang, J., Tang, Z., Lai, X.-C., Qiu, F., Jiang, Y., Liu, Q., Tang, X., Sun, Q., Zhou, Y., Fan, J., et al. (2024) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.4c05316 · Cited by 64
Review of Memristors for In‐Memory Computing and Spiking Neural Networks
Shooshtari, M., Serrano‐Gotarredona, T., & Linares‐Barranco, B. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202500806 · Cited by 74
Pan, J., Kan, H., Liu, Z., Gao, S., Wu, E., Li, Y., & Zhang, C. (2024) · npj Flexible Electronics · doi:10.1038/s41528-024-00356-6 · Cited by 60
“…on buses and memory access but also breaks the traditional von Neumann bottleneck, since each memristor can concurrently perform both memory and logic operations, greatly enhancing chip integration density and signi fi cantly reducing energy consumption between storage and calculation 10 – 14 .”
Recent advances in ferroelectric materials, devices, and in-memory computing applications
Hwang, H., Youn, S., & Kim, H. (2025) · Nano Convergence · doi:10.1186/s40580-025-00520-2 · Cited by 58
Self-citationProgrammable Threshold Logic Implementations in a Memristor Crossbar Array
Youn, S., Lee, J., Kim, S., Park, J., Kim, K., & Kim, H. (2024) · Nano Letters · doi:10.1021/acs.nanolett.3c04073 · Cited by 47
Self-citationTwo-dimensional material-based memristive devices for alternative computing
Panisilvam, J., Lee, H. Y., Byun, S., Fan, D., & Kim, S. (2024) · Nano Convergence · doi:10.1186/s40580-024-00432-7 · Cited by 38
“In this respect, a promising area of research in the future would be the application of nonlinear functions directly in hardware for these structures [178].”
BackgroundSelf-citationMemristor Crossbar Array with Enhanced Device Yield for In-Memory Vector–Matrix Multiplication
Kim, T., Kim, S., Park, J., Youn, S., & Kim, H. (2024) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.4c00199 · Cited by 31
Self-citationPhysical Unclonable Function with 3D Stacked Memristor Crossbar Array Using Self-Differential Pair
Park, J., & Kim, H. (2025) · ACS Nano · doi:10.1021/acsnano.4c18621 · Cited by 30
Self-citationBhatnagar, V., & Kumar, A. (2025) · Energy Storage · doi:10.1002/est2.70272 · Cited by 26
“The VMM operations that use the analogue switching characteristics of the memristor device are useful for neuromorphic computing applications [44].”
Kim, S., Park, K., Hong, K., Kim, T., Park, J., Youn, S., Kim, H., & Choi, W. Y. (2024) · Advanced Materials Technologies · doi:10.1002/admt.202400063 · Cited by 27
Self-citationNeuromorphic Computing with Memcapacitors: Advancements, Challenges, and Future Directions
AbuHamra, N., Khan, M. U., Hassan, E., Al‐Qutayri, M., & Mohammad, B. S. (2025) · Advanced Electronic Materials · doi:10.1002/aelm.202500250 · Cited by 21
Lee, Y., Rhee, H., Kim, G., Cheong, W. H., Kim, D. H., Song, H., Kay, S. N., Lee, J., & Kim, K. M. (2025) · Nature Communications · doi:10.1038/s41467-025-59589-2 · Cited by 19
Self-citationMulti-functional synaptic memristor for neuromorphic pattern recognition and image compression
Sun, H., Li, S., Dong, X., Yang, F., Zhang, X., Chen, J., Zhang, X., Chen, J., Zhao, Y., & Li, Y. (2025) · Materials Today Physics · doi:10.1016/j.mtphys.2025.101684 · Cited by 18
True random number generator using stochastic noise signal of memristor with variation tolerance
Yu, D., Ahn, S., Youn, S., Park, J., & Kim, H. (2024) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2024.115708 · Cited by 16
Self-citationZhao, J., Zhu, Y., Yan, S., Li, G., Liu, R., Zhong, Q., Bian, J., Peng, M., Li, Q., Li, Y., et al. (2025) · Small · doi:10.1002/smll.202500062 · Cited by 15
Zhao, Z., Che, Q., Chen, Q., Wang, K., Zhao, K., Zhang, C., He, H., Wang, X., & Chen, Y. (2024) · Bulletin of the Chemical Society of Japan · doi:10.1093/bulcsj/uoae083 · Cited by 14
Configurable Kernel Map Implementation in Memristor Crossbar for Convolution Neural Network
Kim, G., Youn, S., Park, J., & Kim, H. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202500032 · Cited by 14
Self-citationJang, J., & Gi, S.-G. (2025) · Advanced Materials Technologies · doi:10.1002/admt.202500588 · Cited by 13
Memristor-based in-situ convolutional strategy for accurate braille recognition
Zhang, X., Qin, C., Peng, W., Qin, N., Cheng, E., Wu, J., Fan, Y., Yang, Q., & Chen, H. (2024) · Science China Materials · doi:10.1007/s40843-024-3122-7 · Cited by 13
“Convolutional neural networks (CNNs), which can extract features from large datasets [3], have evolved into mainstream methods in image processing-related fields such as image recognition, image segmentation, and motion detection [4].”
InfluentialHsin, T., Lin, C., Wang, P., Yang, C., & Pai, C. (2025) · Advanced Science · doi:10.1002/advs.202417735 · Cited by 9
Youn, S., Hwang, H., & Kim, H. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202500817 · Cited by 8
Self-citationPark, J., Lee, J., Youn, S., & Kim, H. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202501001 · Cited by 6
Self-citationPark, J., & Kim, H. (2026) · Advanced Functional Materials · doi:10.1002/adfm.202531235 · Cited by 6
Self-citationCoupling-Free Readout Scheme for Memcapacitors With NAND Flash Structure
Ahn, S., Yu, J., Hwang, H., Song, M. S., Yu, D., Hwang, S., Chung, I.-Y., Choi, W. Y., & Kim, H. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3418289 · Cited by 6
“In particular, resistance-based memristors have been a popular choice due to their simple two-terminal structure, fast processing speed, and high cell density [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26].”
BackgroundSelf-citationChang, P.-H., Lin, W.-Y., Huang, Y., Chen, Y.-C., Shih, L., & Chen, J. (2024) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.4c15102 · Cited by 6
Sun, H., Huang, T., Zhang, X., Yang, F., Dong, X., Chen, J., Zhang, X., Chen, J., Zhao, Y., & Li, Y. (2025) · The Journal of Physical Chemistry Letters · doi:10.1021/acs.jpclett.5c02109 · Cited by 5
Youn, S., Hwang, H., & Kim, H. (2025) · Journal of Alloys and Compounds · doi:10.1016/j.jallcom.2025.185701 · Cited by 6
Self-citationZhang, D., Li, R., Zou, X., Meng, H., Liu, Q., Li, Y., & Zhang, C. (2025) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2025.3584318 · Cited by 4
“In time-series data processing, convolutional neural networks (CNNs) are limited due to the lack of memory capacity and insensitivity to temporal dependencies [22], [23], [24].”
Cheng, Z., Wang, H., Guan, Z., Zhu, Z., Shen, S., Yin, Y., & Li, X. (2025) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.5c00740 · Cited by 4
Park, E.-S., Lee, S., & Kim, M. (2025) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.5c11556 · Cited by 3
Ferroelectric memristors optimized in thickness for short-term memory-driven reservoir computing
Park, J., & Kim, S. (2026) · Journal of Materials Chemistry C · doi:10.1039/d5tc03983e · Cited by 3
Self-citationKim, Y. W., An, Y.-J., Kim, S.-H., Kim, Y.-B., Kim, D.-M., Gautam, S. K., Ban, S.-H., Lee, J.-H., Yeom, J.-H., Kim, K., et al. (2026) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.5c02491 · Cited by 3
Self-citationWordline Input Bias Scheme for Neural Network Implementation in 3D-NAND Flash
Hwang, H., Kim, G. H., Yu, D., & Kim, H. (2025) · Biomimetics · doi:10.3390/biomimetics10050318 · Cited by 3
Self-citationA Quantized-Weight-Splitting Method of RRAM Arrays for Neuromorphic Applications
Park, K., Kim, S., Park, J.-H., & Choi, W. Y. (2024) · IEEE Access · doi:10.1109/access.2024.3394253 · Cited by 2
“…can be applied not only to RRAM synapses but also to various two-terminal synaptic devices including ferroelectric tunnel junction (FTJ) and phase change memory (PCM), which represent weights as electrical current, charge, or conductance [30], [31], [32], [33], [34], [35], [36], [37], [38], [39].”
BackgroundSelf-citationXie, Q., Pan, X., Luo, W., Shuai, Y., Wang, Y., Tong, J., Zhao, Z., Wu, C., & Zhang, W. (2024) · Advanced Electronic Materials · doi:10.1002/aelm.202400651 · Cited by 2
Kim, J. I., So, M., Wang, W. J., Kim, T., Jeong, E. S., Sim, K., Park, H., Kim, S., Shin, Y.-G., Choi, J., et al. (2025) · Advanced Science · doi:10.1002/advs.202511801 · Cited by 2
Self-citationPark, J.-H., Kim, S., & Choi, W. Y. (2024) · Advanced Intelligent Systems · doi:10.1002/aisy.202400673 · Cited by 2
Self-citationHassan, E., Ghazalah, S. A., Alshehri, F., & Qadir, S. (2025) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-025-02777-9 · Cited by 1
Noh, T., Kim, C. J., Lee, J. Y., Ku, M., Kim, T., Kang, M., Seong, H. S., Kang, B. S., & Kang, B. S. (2025) · Current Applied Physics · doi:10.1016/j.cap.2025.11.006 · Cited by 1
Self-citationLoizos, M., Rogdakis, K., Chatzimanolis, K., Anagnostou, K., & Kymakis, E. (2026) · Advanced Electronic Materials · doi:10.1002/aelm.202500713 · Cited by 1
WO3−x:N/ZnO:N broadband optoelectronic synapse for object detection in edge computing
Wu, T., Wu, J., Yue, W., Zhang, M., Kim, E.-S., Kim, N., & Li, Y. (2026) · Science China Materials · doi:10.1007/s40843-025-3839-4 · Cited by 1
“Therefore, there is a need to develop a chip for deploying a new neural network system [30–33].”
Youn, S., Hwang, H., & Kim, H. (2026) · Nano Energy · doi:10.1016/j.nanoen.2026.111871 · Cited by 1
Self-citationBorgohain, K. K., Das, U., Dehingia, A., Dutta, R., Patra, S., & Roy, A. (2026) · Journal of Solid State Chemistry · doi:10.1016/j.jssc.2026.126139 · Cited by 1
A Unified Flash Memory Platform for Mode‐Adaptive and Robust AI Computation
Yu, D., Hwang, H., Oh, B., Lee, J., Yu, S., Kim, T., & Kim, H. (2026) · Advanced Science · doi:10.1002/advs.76619 · Cited by 1
Self-citationHardware realization of memory-augmented neural networks using multi-level RRAM arrays
Kim, G., Byun, Y., Kim, S., & Kim, S. (2026) · Nanoscale · doi:10.1039/d6nr02137a · Cited by 1
Self-citationKim, H., Woo, S. Y., & Kim, H. (2024) · Biomimetics · doi:10.3390/biomimetics9060335
Self-citationHigh-Speed Data Writing Strategy Based on Error Weight Mapping
Du, X., Li, Z., Chen, T., Tong, P., & Wang, W. (2024) · International Symposium on Semiconductor and Electronic Technology (ISSET) · doi:10.1109/isset62871.2024.10779816
Neustroev, A. A., Busygin, A. N., & Udovichenko, S. Y. (2025) · Nanobiotechnology Reports · doi:10.1134/s2635167624602705
Lu, H., Chen, Y., & Chen, J.-Y. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202500769
Quasi‐digital memristor with self‐rectifying and synaptic functions in crossbar array architectures
Liu, T., Wang, Q., Liu, Q., Li, J., Xia, Z., Yang, H., Mao, Z., Gou, S., Li, Z., Zhu, C., et al. (2026) · FlexMat. · doi:10.1002/flm2.70073
Liu, Y., Han, T., Ding, J., Lian, H., Yao, L., Xu, Z., Zhang, X., Wang, S., Xia, J., Zhang, T., et al. (2026) · Advanced Science · doi:10.1002/advs.76309
Li, C., Yin, W., & Zheng, Y. (2026) · International Journal of Reasoning-based Intelligent Systems · doi:10.1504/ijris.2026.156715
Unified In‐Memory Hardware Security Enabled by an Entropy Landscape in a Memristor Crossbar
Park, J., & Kim, H. (2026) · Small · doi:10.1002/smll.75550
Self-citationCapacitive in-memory vector–matrix multiplication computing with charge-trap memcapacitors
Yu, J., Hwang, H., Kim, H., & Choi, W. Y. (2026) · Nano Convergence · doi:10.1186/s40580-026-00575-9
“Memristor-based devices, such as resistive switching memory and phase change memory, offer analog programmability and enable conductance-based VMM through current summation during read pulses, but their scalability toward highly integrated three-dimensional arrays remains challenging [14–24].”
Self-citationAnt‐inspired Trimodal Hair‐Like BiVO 4 Memristor for Intelligent Diving‐Assistance Systems
Zhao, K., Lin, R., Zhao, M., Yan, X., & Wang, H. (2026) · Advanced Functional Materials · doi:10.1002/adfm.78920
Charge‐Domain In‐Memory Computing With NAND‐Compatible Ferroelectric Memcapacitor Crossbar
Hwang, H., Yu, J., Youn, S., Lee, D., & Kim, S. (2026) · Small · doi:10.1002/smll.76242
Self-citationElectronic synapses based on ZnTiO₃ memristors for device-aware ANN/SNN neuromorphic computing
Wang, X., LIAO, M., Rong, Z., Lu, Y., Huang, W., Zhou, X., Xu, J., & Liu, Z. (2026) · Surface and Coatings Technology · doi:10.1016/j.surfcoat.2026.134106
Metadata
- Authors
- Park, J., Kim, S., Song, M. S., Youn, S., Kim, K., Kim, T.-H., & Kim, H.
- Affiliations
- Inha University (KR)
- Seoul National University (KR)
- Seoul National University of Science and Technology (KR)
- Hanyang University (KR)
- Corresponding author
- Kim, Hyungjin
- Venue
- ACS Applied Materials & Interfaces
- Publisher
- American Chemical Society (ACS)
- Published
- 2024-01-01
- Volume
- 16
- Issue
- 1
- Pages
- 1054-1065
- ISSN
- 1944-8244 (ISSN-L)
- 1944-8252
- OpenAlex
- W4390478144
- Cited by
- 56
- References
- 64
- Citations by year
- 2026: 16 · 2025: 24 · 2024: 16
- FWCI
- 5.81
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- ACS Applied Materials & Interfaces (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Research Foundation of Korea (2020M3H5A1081111, 2022M3I7A1078544, 2022R1A6A1A03051705, 2021R1C1C1014530)
- Institute for Information and Communications Technology Promotion (IITP-2021-0-02052)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Ferroelectric and Negative Capacitance Devices
- Keywords
- memristor crossbar, convolutional neural network hardware, weight quantization, matrix-vector multiplication, image classification
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.