Design of Ternary Neural Network With 3-D Vertical RRAM Array
The DOI matches its Crossref record; 27 of the 29 references it cites were found in a registry or an index.
Abstract
Recently, 2-D cross-point array of resistive random access memory (RRAM) has been proposed for implementing the weighted sum and weight update operations to accelerate the neuro-inspired learning algorithms on chip. This paper aims to extend such 2-D cross-point array to 3-D vertical array for storing and computing the large-scale weight matrices in the neural network. Considering the fabrication and 3-D integration of analog synapses (i.e., multilevel RRAM devices) are premature at this stage, we propose using today's available digital or binary RRAM devices for implementing a ternary neural network, which aggressively reduces the weight precision to ternary levels (+1, 0,-1) for the weighted sum in both feedforward and backward inference, while the multiple 3-D layers could serve for accumulating the small errors in a higher precision format for weight update. Compared to the 2-D implementation, the proposed 3-D vertical implementation shows larger read/write margin for weighted sum/weight update, smaller latency, and energy consumption for weight update. This paper demonstrates the attractiveness for building a monolithic 3-D neuromorphic hardware platform.
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
“Previous work on 3D RRAM-based nvCIMs has been limited only to the memory array, and effective 3D computing systems that include peripheral circuits are lacking 20-31 .”
From A computing-in-memory macro based on three-dimensional resistive random-access memory (2022) · cites this paper as [27]
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
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 · cited 1×
“where f denotes the sigmoid activation function [22] with limited s-bit precision truncation.”
BackgroundTraining and operation of an integrated neuromorphic network based on metal-oxide memristors
Prezioso, M., Merrikh-Bayat, F., Hoskins, B. D., Adam, G. C., Likharev, K. K., & Strukov, D. B. (2015) · Nature · doi:10.1038/nature14441 · Cited by 3,001 · cited 1×
“, 12 × 12 TiOx/Al2O3 crossbar [5]) to medium-scale arrays (e.”
BinaryConnect: Training Deep Neural Networks with binary weights during propagations
Courbariaux, M., Bengio, Y., & David, J.-P. (2015) · arXiv · doi:10.48550/arxiv.1511.00363 · Cited by 2,180 · cited 2×
“Such slight accuracy degradation is within the expectation [16], [17].”
BackgroundSeok, J. Y., Song, S. J., Yoon, J. H., Yoon, K. J., Park, T. H., Kwon, D. E., Lim, H., Kim, G. H., Jeong, D. S., & Hwang, C. S. (2014) · Advanced Functional Materials · doi:10.1002/adfm.201303520 · Cited by 383 · cited 1×
“of necessary interconnection lines [15].”
Yu, S., Chen, H.-Y., Gao, B., Kang, J., & Wong, H.-S. P. (2013) · ACS Nano · doi:10.1021/nn305510u · Cited by 339
Self-citationSuri, M., Bichler, O., Querlioz, D., Cueto, O., Perniola, L., Sousa, V., Vuillaume, D., Gamrat, C., & DeSalvo, B. (2011) · 2011 International Electron Devices Meeting · doi:10.1109/iedm.2011.6131488 · Cited by 314 · cited 4×
“In the recent years, emerging nonvolatile memory such as phase-change memory (PCM) [2] and resistive random access memory (RRAM) [3], [4] based “analog” synapses that exploit the multilevel states have been demonstrated at single device level.”
BackgroundInfluentialBackpropagation for energy-efficient neuromorphic computing
Esser, S. K., Appuswamy, R., Merolla, P. A., Arthur, J. V., & Modha, D. S. (2015) · neural information processing systems · Cited by 281
Chen, A. (2013) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2013.2246791 · Cited by 230
Yu, S., Guan, X., & Wong, H.-S. P. (2012) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2012.2184544 · Cited by 224 · cited 1×
“To set up appropriate values for VW and the criterion for the write margin, the switching voltage distribution in an array should be considered as RRAM devices are well known for the switching parameter variability [26].”
BackgroundSelf-citation
Research works printed but not linked: 3
- hsu, 2013, 3D Vertical TaOx/TiO2 RRAM with over 103 selfrectifying ratio and sub- $\mu$ a operating current, IEDM Tech Dig, 10.4.1
- lecun, 2012, The MNIST Database of Handwritten Digits
- chen, 2012, HfOx based vertical RRAM for cost-effective 3D cross-point architecture without cell selector, IEDM Tech Dig, 20.7.1
Cited by
Neuro-Inspired Computing With Emerging Nonvolatile Memorys
Yu, S. (2018) · Proceedings of the IEEE · doi:10.1109/jproc.2018.2790840 · Cited by 1,170
“hybrid precision requirement for the weight propagation and weight update [127].”
BackgroundSelf-citationThree-dimensional memristor circuits as complex neural networks
Lin, P., Li, C., Wang, Z., Li, Y., Jiang, H., Song, W., Rao, M., Zhuo, Y., Upadhyay, N. K., Barnell, M., et al. (2020) · Nature Electronics · doi:10.1038/s41928-020-0397-9 · Cited by 375
“The 3D design provides a high degree of functional complexity, which goes beyond the purpose of increasing packing density in conventional 3D architectures 20–26 and opens a critical avenue to using memristors for neuromorphic computing.”
BackgroundRecent progress in analog memory-based accelerators for deep learning
Tsai, H., Ambrogio, S., Narayanan, P., Shelby, R. M., & Burr, G. W. (2018) · Journal of Physics D: Applied Physics · doi:10.1088/1361-6463/aac8a5 · Cited by 225
Analog architectures for neural network acceleration based on non-volatile memory
Xiao, T. P., Bennett, C. H., Feinberg, B., Agarwal, S., & Marinella, M. J. (2020) · Applied Physics Reviews · doi:10.1063/1.5143815 · Cited by 193
A computing-in-memory macro based on three-dimensional resistive random-access memory
Huo, Q., Yang, Y., Wang, Y., Lei, D., Fu, X., Ren, Q., Xu, X., Luo, Q., Xing, G., Chen, C., et al. (2022) · Nature Electronics · doi:10.1038/s41928-022-00795-x · Cited by 159
“When adopting conventional 3D VRRAMs, using WL or SL as the input cannot guarantee that all the devices participate in VMM operations in one computing cycle 27,29 .”
MethodEnergy Consumption Prediction Using Machine Learning; A ReviewPreprint
Mosavi, A. H., & Bahmani, A. (2019) · Preprints.org · doi:10.20944/preprints201903.0131.v1 · Cited by 125
Bio‐Inspired 3D Artificial Neuromorphic Circuits
Liu, X., Wang, F., Su, J., Zhou, Y., & Ramakrishna, S. (2022) · Advanced Functional Materials · doi:10.1002/adfm.202113050 · Cited by 91
Emerging memory technologies for neuromorphic computing
Kim, C.-H., Lim, S., Woo, S. Y., Kang, W.-M., Seo, Y.-T., Lee, S., Lee, S., Kwon, D., Oh, S., Noh, Y., et al. (2018) · Nanotechnology · doi:10.1088/1361-6528/aae975 · Cited by 91
Phase Change Random Access Memory for Neuro‐Inspired Computing
Wang, Q., Niu, G., Ren, W., Wang, R., Chen, X.-G., Li, X., Ye, Z., Xie, Y., Song, S., & Song, Z. (2021) · Advanced Electronic Materials · doi:10.1002/aelm.202001241 · Cited by 74
Byun, K., Choi, I., Kwon, S., Kim, Y., Kang, D., Cho, Y. W., Yoon, S. K., & Kim, S. (2022) · Advanced Materials Technologies · doi:10.1002/admt.202200884 · Cited by 68
A Three-Dimensional Neuromorphic Photosensor Array for Nonvolatile In-Sensor Computing
Wang, Y., Cai, Y., Wang, F., Yang, J., Yan, T., Li, S., Wu, Z., Zhan, X., Xu, K., He, J., et al. (2023) · Nano Letters · doi:10.1021/acs.nanolett.3c00899 · Cited by 59
Design implementations of ternary logic systems: A critical review
Zahoor, F., Jaber, R. A., Bature, U. I., Sharma, T., Bashir, F. I., Abbas, H., Alzahrani, A., Gupta, S., & Hanif, M. (2024) · Results in Engineering · doi:10.1016/j.rineng.2024.102761 · Cited by 51
Memristor Based Binary Convolutional Neural Network Architecture With Configurable Neurons
Huang, L., Diao, J., Nie, H., Wang, W., Li, Z., Li, Q., & Liu, H. (2021) · Frontiers in Neuroscience · doi:10.3389/fnins.2021.639526 · Cited by 45
“During the process of simulation, the resistance in memristor array is randomly changed to be GHRS or GLRS for emulating the defect of S-AH or S-A-L (Liu et al., 2017).”
MethodSelf-citationCross-point Resistive Memory: Nonideal Properties and Solutions
Wang, C., Feng, D., Tong, W., Liu, J., Li, Z., Chang, J., Zhang, Y., Wu, B., Xu, J., Zhao, W. J., et al. (2019) · ACM Transactions on Design Automation of Electronic Systems · doi:10.1145/3325067 · Cited by 40
“Three-dimensional arrays provide abundant operation forms [116].”
BackgroundSelf-citationPerspective: Uniform switching of artificial synapses for large-scale neuromorphic arrays
Tan, S. H., Lin, P., Yeon, H., Choi, S., Park, Y., & Kim, J. (2018) · APL Materials · doi:10.1063/1.5049137 · Cited by 33
A Reconfigurable 4T2R ReRAM Computing In-Memory Macro for Efficient Edge Applications
Chen, Y., Lu, L., Kim, B., & Kim, T. T.-H. (2021) · IEEE Open Journal of Circuits and Systems · doi:10.1109/ojcas.2020.3042550 · Cited by 32
“For example, a BITW ReRAM-based multilayer perception (MLP) in [29] achieves a 95% classification accuracy on MNIST even with a 20% standard deviation in ReRAM resistances.”
MethodA Logic Compatible 4T Dual Embedded DRAM Array for In-Memory Computation of Deep Neural Networks
Yoo, T., Kim, H., Chen, Q., Kim, T. T.-H., & Kim, B. (2019) · International Symposium on Low Power Electronics and Design (ISLPED) · doi:10.1109/islped.2019.8824826 · Cited by 32
A Reliable and Energy-Efficient Nonvolatile Ternary Memory Based on Hybrid FinFET/RRAM Technology
Yousefi, A., Eslami, N., & Moaiyeri, M. H. (2022) · IEEE Access · doi:10.1109/access.2022.3211562 · Cited by 30
“The RRAM arrays can also be used for implementing efficient ternary neural networks and ternary content-addressable memory (TCAM) [17], [18].”
MethodMetal–Organic Frameworks–Based Memristors: Materials, Devices, and Applications
Shu, F., Chen, X., Yu, Z., Gao, P., & Liu, G. (2022) · Molecules · doi:10.3390/molecules27248888 · Cited by 28
Khodayari, F., Amirany, A., Moaiyeri, M. H., & Jafari, K. (2023) · IEEE Transactions on Magnetics · doi:10.1109/tmag.2022.3233891 · Cited by 27
“M ANY researchers have devoted themselves to the design of ternary logic circuits in recent years [1], [2], [3].”
BackgroundDu, Y., Jing, L., Fang, H., Chen, H., Cai, Y., Wang, R., Zhang, J., & Ji, Z. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.3002736 · Cited by 26
Time‐Efficient Stateful Dual‐Bit‐Memristor Logic
Xu, N., Fang, L., Kim, K. M., & Hwang, C. S. (2019) · physica status solidi (RRL) - Rapid Research Letters · doi:10.1002/pssr.201900033 · Cited by 24
A Multi-Bit Neuromorphic Weight Cell Using Ferroelectric FETs, suitable for SoC Integration
Obradovic, B. J., Rakshit, T., Hatcher, R., Kittl, J. A., Sengupta, R., Hong, J. G., & Rodder, M. S. (2018) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2018.2817628 · Cited by 22
2D van der Waals Vertical Heterojunction Transistors for Ternary Neural Networks
Li, Z., Huang, X., Xu, L., Peng, Z., Yu, X., Shi, W., He, X., Meng, X., Yang, D.-H., Tong, L., et al. (2023) · Nano Letters · doi:10.1021/acs.nanolett.3c03553 · Cited by 21
Self-citationReSe2-Based RRAM and Circuit-Level Model for Neuromorphic Computing
Huang, Y., Gu, Y., Wu, X., Ge, R., Chang, Y., Wang, X., Zhang, J., Akinwande, D., & Lee, J. C. (2021) · Frontiers in Nanotechnology · doi:10.3389/fnano.2021.782836 · Cited by 21
“The operation of RRAM devices is based on the non-volatile resistive switching (NVRS) phenomenon of the active layer, typically a bulk metal-oxide fi lm, such as SiO 2 or HfO 2 , in a metal-insulator-metal (MIM) stacking structure (Chang et al., 2016; Lin et al., 2017).”
BackgroundIn Situ Learning in Hardware Compatible Multilayer Memristive Spiking Neural Network
Li, J., Xu, H., Sun, S., Li, N., Li, Q., Li, Z., & Liu, H. (2021) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2021.3049487 · Cited by 20
“Based on Kirchhoff’s law, some published literatures apply memristor crossbar array to realize multiply accumulate calculation (MAC) in CNNs, avoiding data movement between the processing unit and the memory [8]–[10].”
MethodSelf-citationWoo, J., & Yu, S. (2018) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2018.2875937 · Cited by 19
Self-citationSun, W., Choi, S., Kim, B., & Park, J. (2019) · Materials · doi:10.3390/ma12203451 · Cited by 19
“For example, the four-layer 3D RRAM integrated with FinFET (Fin Field-Effect Transistor) was developed for brain-inspired computing and in-memory computing [11], and 3D vertical array of RRAM was proposed for storing and computing large-scale weight matrices in the neural network [12].”
MethodLeveraging 3D Vertical RRAM to Developing Neuromorphic Architecture for Pattern Classification
Kim, B., & Li, H. H. (2020) · IEEE Computer Society Annual Symposium on VLSI · doi:10.1109/isvlsi49217.2020.00054 · Cited by 17
“In particular, using multiple binary RRAM devices to represent one synaptic weight has been adopted in many designs [8], [10].”
MethodTutorial: Fabrication and three-dimensional integration of nanoscale memristive devices and arrays
Lin, P., & Xia, Q. (2018) · Journal of Applied Physics · doi:10.1063/1.5038109 · Cited by 16
Implementation of Ternary Weights With Resistive RAM Using a Single Sense Operation Per Synapse
Laborieux, A., Bocquet, M., Hirtzlin, T., Klein, J., Nowak, E., Vianello, E., Portal, J., & Querlioz, D. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.3031627 · Cited by 16
Ryu, J., & Kim, S. (2020) · Materials · doi:10.3390/ma13184201 · Cited by 16
Modeling and Simulation of Resistive Random Access Memory With Graphene Electrode
Xie, H., Chen, W., Zhang, S., Zhu, G., Khaliq, A., Hu, J. X., & Yin, W. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.2965182 · Cited by 16
Implication of unsafe writing on the MAGIC NOR gate
Zhu, X. J., Long, H., Li, Z., Diao, J., Liu, H., Li, N., & Xu, H. (2020) · Microelectronics Journal · doi:10.1016/j.mejo.2020.104866 · Cited by 15
Self-citationResistive switching memory for high density storage and computing*
Xu, X., Luo, Q., Gong, T., Lv, H., Liu, Q., & Liu, M. (2021) · Chinese Physics B · doi:10.1088/1674-1056/abe0c4 · Cited by 15
Graphene-based 3D XNOR-VRRAM with ternary precision for neuromorphic computing
Alimkhanuly, B., Sohn, J., Chang, I., & Lee, S. (2021) · npj 2D Materials and Applications · doi:10.1038/s41699-021-00236-x · Cited by 15
“Alternatively, the WPs and SLs can be combined to perform the VMM operation 23 , in which the selected and unselected WPs are biased to V R and 0 V, respectively, and selected BLs should be grounded (Fig.”
BackgroundSelf-controlled multilevel writing architecture for fast training in neuromorphic RRAM applications
Garćıa-Redondo, F., & López‐Vallejo, M. (2018) · Nanotechnology · doi:10.1088/1361-6528/aad2fa · Cited by 13
“For this reason, some researchers have presented alternatives to MLC synapses, involving Ternary NN [ 10 ] .”
BackgroundChang, Y., Fowler, B. W., Chen, Y., Lin, C.-Y., Xu, G., Huang, H., Chen, J., Kim, S., Li, Y., & Lee, J. C. (2018) · Journal of Materials Chemistry C · doi:10.1039/c8tc04355h · Cited by 11
Quaternary synapses network for memristor-based spiking convolutional neural networks
Sun, S., Li, J., Li, Z., Liu, H., Liu, H., & Li, Q. (2019) · IEICE Electronics Express · doi:10.1587/elex.16.20190004 · Cited by 10
“Although, the convolutional neural networks are competent in tolerating the random e ff ects, such as the device variations or circuit noise [13], they may signi fi cantly degrade the recognition accuracy rate [14, 15].”
BackgroundSelf-citationEnhanced Spiking Neural Network with forgetting phenomenon based on electronic synaptic devices
Li, J., Xu, H. F., Sun, S., Liu, S., Li, N., Li, Q., Liu, H., & Li, Z. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.09.030 · Cited by 10
Self-citationMemristor-based vector neural network architecture*
Liu, H., Chen, C., Zhu, X. J., Sun, S., Li, Q., & Li, Z. (2019) · Chinese Physics B · doi:10.1088/1674-1056/ab65b5 · Cited by 9
Self-citationSolution to alleviate the impact of line resistance on the crossbar array
Zhu, X. J., Li, Z., Liu, H., Li, Q., Liu, S., Li, N., & Xu, H. (2020) · IET Circuits Devices & Systems · doi:10.1049/iet-cds.2019.0313 · Cited by 9
“And the RRAM device in SPICE [27] is modelled as a resistor with a tunable resistance between OFF-state resistance ( R off ) and ON-state resistance ( R on ).”
MethodSelf-citationA method of generating random bits by using electronic bipolar memristor*
Yang, B., Xu, N., Zhou, E., Li, Z., Li, C., Yi, P., & Fang, L. (2020) · Chinese Physics B · doi:10.1088/1674-1056/ab77fd · Cited by 9
Self-citationEEG Signal Epilepsy Detection System Based on Convolutional Neural Network and Memristor Array
Zhang, J., Xu, H. F., Long, H., Hou, K., Li, Z., Yang, Z., Wang, Y., & Li, Q. (2021) · International Conference on Electronics Technology (ICET) · doi:10.1109/icet51757.2021.9450952 · Cited by 9
“The raw EEG signal contains large quantity of noise and needs to be pre-processed [7-10] .”
BackgroundSelf-citationKim, Y., Kim, H., Jeon, S., Kim, H. W., Hong, E., Kim, N., Choi, H., Park, H., Jeong, J., Lee, D., et al. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3408784 · Cited by 8
“To implement these systems in hardware, researchers have utilized a cross-point array architecture with memory devices for electronic synapses [3], [4], [5].”
MethodLRS retention fail based on joule heating effect in InGaZnO resistive-switching random access memory
Jang, J. T., Ahn, G., Choi, S., Kim, D. M., Kim, H., & Kim, D. H. (2020) · Applied Physics Express · doi:10.35848/1882-0786/ab88c1 · Cited by 8
Evolutionary Learning of Binary Neural Network Using a TaO x Memristor via Stochastic Stateful Logic
Kim, D. H., Kim, Y. S., Cheong, W. H., Song, H., Rhee, H., Kay, S. N., Han, J., & Kim, K. M. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200058 · Cited by 8
Pipelined Memristive Analog-to-Digital Converter With Self-Adaptive Weight Tuning
Wang, W., Li, Z., Fu, X., Wang, Y., & Li, Q. (2022) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2022.3221083 · Cited by 8
Self-citationIn-situ learning in multilayer locally-connected memristive spiking neural network
Li, J., Xu, H., Sun, S., Li, Z., Li, Q., Liu, H., & Li, N. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.08.011 · Cited by 7
Self-citationSpintronic Memristor-Based Binarized Ensemble Convolutional Neural Network Architectures
Tchendjou, G. T., Danouchi, K., Prenat, G., & Anghel, L. (2022) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2022.3213612 · Cited by 7
“For a full evaluation of the energy consumption of the entire ensemble network architecture, we consider the estimation methodology presented in papers [10] and [54].”
MethodCAMEM: A Computationally-Efficient and Accurate Memristive Model With Experimental Verification
Hajri, B., Mansour, M. M., Chehab, A., & Aziza, H. (2019) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2019.2945985 · Cited by 6
“capturing only the resistance characteristic is highly efficient for large scale arrays [7].”
BackgroundLee, W. J., Kim, B., Koo, M., & Kim, Y. (2024) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.3c01680 · Cited by 5
Alimkhanuly, B., Lee, M., Lee, S., Devnath, A., Bae, J., Choi, J., Patil, S. V., Sim, T., Moon, S., Lee, G., et al. (2025) · Advanced Materials · doi:10.1002/adma.202509367 · Cited by 5
Binary Memristive Synapse Based Vector Neural Network Architecture and Its Application
Liu, H., Sun, S., Liu, J., Li, Q., Diao, J., & Li, Z. (2020) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2020.3015337 · Cited by 5
“The training process constitutes of continuous adjustment of the synaptic weights based on the training We have employed the sigmoid as the activation function, denoted as f , and adopt the method of quantization aware training to obtain the tri-state weights [13], [16].”
BackgroundMethodInfluentialSelf-citationZhao, G., Jin, X., Ye, H., Peng, Y., Liu, W., Yin, N., Chen, W., Chen, J., Li, X., & Yu, Z. (2022) · IEICE Electronics Express · doi:10.1587/elex.19.20220399 · Cited by 4
Castellani-Coulié, K., Aziza, H., & Rahajandraibe, W. (2023) · Journal of Electronic Testing · doi:10.1007/s10836-023-06068-5 · Cited by 4
“Then, the worst structure would correspond to 3D stacking, where NMOS and PMOS transistors could be physically close [24].”
BackgroundAccelerating 3D vertical resistive memories with opportunistic write latency reduction
Wen, W., Zhang, Y., & Yang, J. (2020) · doi:10.1145/3400302.3415677 · Cited by 3
“Many recent work [2, 5, 8, 12, 14, 17, 18, 21] proposed to exploit the feature of analog current aggregation in ReRAM crossbars for accelerating multiply-add calculations between matrices and vectors.”
BackgroundMethodCases Study of Inputs Split Based Calibration Method for RRAM Crossbar
Sun, S., Xu, H., Li, J., Li, Z., Sun, Y., Li, Q., & Liu, H. (2019) · IEEE Access · doi:10.1109/access.2019.2944417 · Cited by 3
“Most of works focused on the memristor fabrication reliability as electronic synapses [21], [22], and some works proposed solutions to neuromorphic computing with few levels [23], [24] or damaged device [25].”
BackgroundSelf-citationEfficient Approximate Ternary Multipliers for Emerging Nanodevices
Krishna, L. H., Bodapati, S., & Sridharan, K. (2025) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2025.3648734 · Cited by 2
“Ternary neural networks have also been studied recently [19], [20], [21].”
Kim, Y., Choi, H., & Woo, J. (2024) · IEEE Silicon Nanoelectronics Workshop · doi:10.1109/snw63608.2024.10639252 · Cited by 2
Adaptive image recognition circuit based on CMOS-based memristor
Xie, S., Gong, J., & Mao, X. (2022) · IEICE Electronics Express · doi:10.1587/elex.19.20220440 · Cited by 2
SCMOS: Series-Connected Memristor-only Stateful Logic
Li, Z., Zhu, X. J., Li, N., & Long, H. (2020) · IEEE International Conference on Solid-State and Integrated Circuit Technology · doi:10.1109/icsict49897.2020.9278249 · Cited by 1
Self-citationHigh throughput N-modular redundancy for error correction design of memristive stateful logic
Zhu, X. J., Xu, H., Yang, W., Li, Z., Liu, H., Liu, S., Wang, Y., & Long, H. (2022) · Chinese Physics B · doi:10.1088/1674-1056/ac8f38 · Cited by 1
Self-citationPerformance comparison of planar and cylindrical ferroelectric tunnel junctions
Yirong, G., Li, J., & Chang, P. (2023) · Japanese Journal of Applied Physics · doi:10.35848/1347-4065/ad184e · Cited by 1
Kim, Y., Jeon, S., Lim, S., & Woo, J. (2025) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps48204.2025.10983653 · Cited by 1
“To implement such systems at the hardware level, non-volatile memory devices capable of analog memory storage are required to replicate the function of synapses [3], [4].”
Bio-inspired continual learning and credit assignment for neuromorphic computing
Laborieux, A. (2021) · theses.fr (ABES) · doi:10.70675/800314bdz09a6z46fdz87f5z3a5795d98157 · Cited by 1
Multiphysics Simulation of Multifilamentary Conduction Resistive Random Access Memory
Xie, H., Zhang, S., Zhu, G., Yin, W., Chen, W., & Hu, J. X. (2019) · IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization · doi:10.1109/nemo.2019.8853829
Long, H., Zhu, X. J., Li, Z., Diao, J., Liu, H., & Li, Q. (2019) · doi:10.1109/iwofc48002.2019.9078439
“However, so far, no corresponding results have been proposed that describing MAGIC NAND gate within a crossbar array [3], [4], [5], [6], and it just tested NAND function with a separate circuit.”
ResultSelf-citationInstability changes the MAGIC NAND gate to the NOR gate
Zhu, X. J., Long, H., Li, Z., Xu, H., Liu, H., & Li, Q. (2019) · doi:10.1109/iwofc48002.2019.9078441
“3 shows an example of relationship between switching probability and instability of threshold voltage for a certain pulse [9].”
BackgroundSelf-citationImproving Performance and Endurance for Crossbar Resistive Memory
Wen, W. (2020) · D-Scholarship@Pitt (University of Pittsburgh)
MoS2-Based RRAM with Asymmetric Contacts
Yan, H., & Ma, Z. (2025) · IEEE International Conference on Electron Devices and Solid-State Circuits (EDSSC) · doi:10.1109/edssc64492.2025.11182720
“Previous studies have primarily focused on device area scaling, electrode material selection, and enhancing material quality to reduce the operation power and improve device reliability [6]–[8].”
Metadata
- Authors
- Li, Z., Chen, P.-Y., Xu, H., & Yu, S.
- Affiliations
- National University of Defense Technology (CN)
- Arizona State University (US)
- Venue
- IEEE Transactions on Electron Devices
- Published
- 2017-06
- Volume
- 64
- Issue
- 6
- Pages
- 2721-2727
- ISSN
- 0018-9383 (ISSN-L)
- 1557-9646
- OpenAlex
- W2610452982
- Cited by
- 70
- References
- 29
- Citations by year
- 2025: 4 · 2024: 4 · 2023: 5 · 2022: 11 · 2021: 9 · 2020: 15 · 2019: 13 · 2018: 9
- FWCI
- 4.21
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Electron Devices (publishedVersion)
- Funders
- National Natural Science Foundation of China Program (61471377)
- National Science Foundation (NSF-CCF-1552687)
- NUDT Outstanding Graduate Student Innovation Fund (B150403)
- 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
- ternary neural networks, resistive random access memory, cross-point array, weight updating, neuromorphic hardware, monolithic 3D integration, analog synapses, energy-efficient inference
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.