Neuro-Inspired Computing With Emerging Nonvolatile Memorys
The DOI matches its Crossref record; 134 of the 141 references it cites were found in a registry or an index.
Abstract
This comprehensive review summarizes state of the art, challenges, and prospects of the neuro-inspired computing with emerging nonvolatile memory devices. First, we discuss the demand for developing neuro-inspired architecture beyond today's von-Neumann architecture. Second, we summarize the various approaches to designing the neuromorphic hardware (digital versus analog, spiking versus nonspiking, online training versus offline training) and discuss why emerging nonvolatile memory is attractive for implementing the synapses in the neural network. Then, we discuss the desired device characteristics of the synaptic devices (e.g., multilevel states, weight update nonlinearity/asymmetry, variation/noise), and survey a few representative material systems and device prototypes reported in the literature that show the analog conductance tuning. These candidates include phase change memory, resistive memory, ferroelectric memory, floating-gate transistors, etc. Next, we introduce the crossbar array architecture to accelerate the weighted sum and weight update operations that are commonly used in the neuro-inspired machine learning algorithms, and review the recent progresses of array-level experimental demonstrations for pattern recognition tasks. In addition, we discuss the peripheral neuron circuit design issues and present a device-circuit-algorithm codesign methodology to evaluate the impact of nonideal device effects on the system-level performance (e.g., learning accuracy). Finally, we give an outlook on the customization of the learning algorithms for efficient hardware implementation.
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
“To implement neural network computing hardware with a high learning accuracy, a low nonlinearity of synaptic weights with repetitive input pulses and a large on/off synapse conductance ratio are required 36 .”
From Enhanced analog synaptic behavior of SiNx/a-Si bilayer memristors through Ge implantation (2020) · cites this paper as [36]
In the citing paper
“The achieved learning accuracy is very comparable to the state-of-the-art results in the literature 36,46,49 .”
From Enhanced analog synaptic behavior of SiNx/a-Si bilayer memristors through Ge implantation (2020) · cites this paper as [36]
In the citing paper
“A promising solution to this problem is the use of memristor-based systems, which can drastically reduce the energy consumption of AI5,6, making it even conceivable to create self-powered edge AI systems that do not require batteries and can instead harvest energy from the environment.”
From Powering AI at the edge: A robust, memristor-based binarized neural network with near-memory computing and miniaturized solar cell (2024) · cites this paper as [5]
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 · cited 1×
“For example, Microsoft’s Res-Net (which won the ImageNet 2015 image classification competition [5]) has more than 100 of layers [6].”
MethodVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 1×
“The adoption of low-precision weight is most suitable for the feedforward inference stage of the CNN models, including LeNet5 [127], AlexNet [3], and VGG [128] in an order of complexity.”
MethodImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 2×
“In particular, convolutional neural network (CNN) [3] and recurrent neural network (RNN) [4] algorithms and their variants have proved their efficacy in a wide range of image, video, speech, and biomedical applications.”
BackgroundMethodLeCun, 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×
“The adoption of low-precision weight is most suitable for the feedforward inference stage of the CNN models, including LeNet5 [127], AlexNet [3], and VGG [128] in an order of complexity.”
MethodMastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612 · cited 1×
“, Google’s AlphaGo [1]).”
Speech recognition with deep recurrent neural networks
Graves, A., Mohamed, A.-R., & Hinton, G. (2013) · 2013 IEEE International Conference on Acoustics, Speech and Signal Processing · doi:10.1109/icassp.2013.6638947 · Cited by 8,893 · cited 1×
“In particular, convolutional neural network (CNN) [3] and recurrent neural network (RNN) [4] algorithms and their variants have proved their efficacy in a wide range of image, video, speech, and biomedical applications.”
BackgroundRedox‐Based Resistive Switching Memories – Nanoionic Mechanisms, Prospects, and Challenges
Waser, R., Dittmann, R., Staikov, G., & Szot, K. (2009) · Advanced Materials · doi:10.1002/adma.200900375 · Cited by 5,080 · cited 1×
“RRAM has two subcategories [44]: one is anion-based oxide random access memory (OxRAM) and the other one is cation-based conductive bridge random access memory (CBRAM) [45].”
BackgroundIn-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537
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 · cited 1×
“A similar work that constraints the weights and neurons to be (+1, −1), namely XNOR-Net [130], has been demonstrated for the classification of ImageNet data set with slight degradation of accuracy.”
Background
Research works printed but not linked: 9
- jouppi, 2016, Google Supercharges Machine Learning Tasks with TPU Custom Chip
- manan, 2012, CBRAM devices as binary synapses for low-power stochastic neuromorphic systems: Auditory (cochlea) and visual (retina) cognitive processing applications, IEDM Tech Dig
- 0, The MNIST dataset of handwritten digits
- 0, Image
- liu, 2013, A 130.7 mm $^{2}~2$ -layer 32 Gb ReRAM memory device in 24nm technology, IEEE Int Solid-State Circuits Conf (ISSCC) Dig Tech Papers, 31.6.1
- 0, Theano
- govoreanu, 2011, $10\times 10$ nm2 Hf/HfOx crossbar resistive RAM with excellent performance, reliability and low-energy operation, IEDM Tech Dig
- 0, Yale Face Database
- 0, NeuroSim
Cited by
Items 601–700 of 1,001
Kwon, O., Lee, D. H., Oh, S., Yoon, J., Kim, H., Ahn, J., Park, W., Kim, H. S., & Cho, B. (2025) · Advanced Functional Materials · doi:10.1002/adfm.202513449 · Cited by 13
All‐Solid‐State Electrolyte‐Gated Synaptic Transistor Array for Deep Learning Hardware Accelerators
Yin, Z., Hao, D., Ci, R., Miao, G., Wang, Y., Yao, D., Liu, A., & Shan, F. (2025) · Advanced Functional Materials · doi:10.1002/adfm.202425471 · Cited by 12
Molecular electronic devices based on atomic manufacturing methods
Yao, C., Li, Y., Zhang, H., Wang, D., Wang, J., Wang, X., Li, X., Liu, J., & Hong, W. (2025) · Microsystems & Nanoengineering · doi:10.1038/s41378-025-01037-8 · Cited by 11
“Their compatibility with self-assembly strategies offers a unique technological pathway toward developing memory-computing integrated chips that combine high density, superior computing power, and low energy consumption 317 .”
Inference engine benchmarking across technological platforms from CMOS to RRAM
Peng, X., Kim, M., Sun, X., Yin, S., Rakshit, T., Hatcher, R., Kittl, J. A., Seo, J.-S., & Yu, S. (2019) · International Symposium on Memory Systems (MEMSYS) · doi:10.1145/3357526.3357566 · Cited by 13
“To achieve even higher integration density, the emerging non-volatile memory (eNVM) devices (with cell size 4~12F(2)) are proposed for the PIM architectures [7].”
Self-citationKang, S., Lee, J., Kang, M., & Song, Y. (2020) · Electronics · doi:10.3390/electronics9081268 · Cited by 11
RetractedOperation and Design of Ferroelectric FETs for a BEOL Compatible Device Implementation
Lizzit, D., & Esseni, D. (2021) · European Solid-State Device Research Conference · doi:10.1109/essderc53440.2021.9631764 · Cited by 11
“applications in crossbar arrays for artificial deep neural networks [3], and in hybrid memristive-CMOS circuits for neuromorphic computing [4, 5].”
BackgroundBaroni, A., Glukhov, A., Pérez, E., Wenger, C., Calore, E., Schifano, S. F., Olivo, P., Ielmini, D., & Zambelli, C. (2022) · Frontiers in Neuroscience · doi:10.3389/fnins.2022.932270 · Cited by 11
“RRAM enables massive IMC parallelism in performing the matrix-vector-multiplication (MVM) in one computational step (i.e., one clock cycle) via the physical laws of Ohm and Kirchhoff in the analog domain (Hu et al., 2018; Ielmini and Wong, 2018; Ma et al., 2018; Yu, 2018).”
BackgroundMethodGoda, A., Matsui, C., & Takeuchi, K. (2022) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2022.3206317 · Cited by 11
“Nonvolatile memory (NVM) technologies have been actively investigated for both synaptic devices and neuron devices [4], [5], [6], [7].”
BackgroundWu, Y., Cai, L., Miao, L., Wang, Z., & Lü, Y. (2023) · Materials Research Bulletin · doi:10.1016/j.materresbull.2023.112152 · Cited by 11
Short-Term Memory Characteristics of IGZO-Based Three-Terminal Devices
Pyo, J., Bae, J., Kim, S., & Cho, S. (2023) · Materials · doi:10.3390/ma16031249 · Cited by 11
“The synapses update the weights by reacting to the signals from other neurons [21].”
BackgroundPhysical neural networks with self-learning capabilities
Yu, W., Guo, H., Xiao, J., & Shen, J. (2024) · Science China Physics Mechanics and Astronomy · doi:10.1007/s11433-024-2403-x · Cited by 11
Xiang, Y., Huang, P. C., Han, R., Zhou, Z., Shu, Q., Su, Z., Hu, H., Liu, L., Liu, Y., Liu, X., et al. (2019) · Advanced Materials Technologies · doi:10.1002/admt.201800720 · Cited by 11
Bennett, C. H., Parmar, V., Calvet, L. E., Klein, J., Suri, M., Marinella, M. J., & Querlioz, D. (2019) · IEEE Access · doi:10.1109/access.2019.2920076 · Cited by 12
“In the first case, SET pulses have more adaptive power (large 1G) when the device is low conductance and less adaptive power (small 1G) as the device approaches its maximum conductance; inversely, RESET pulses are more effective when the device is in a high conductance state [37].”
BackgroundLiu, Y., Gao, B., Xu, F., Zhang, W., Xi, Y., Tang, J., & Qian, H. (2021) · IEEE Electron Devices Technology & Manufacturing Conference (EDTM) · doi:10.1109/edtm50988.2021.9421000 · Cited by 11
“Filamentary analog RRAM has multiple weak conductive filaments (CFs) [7][8].”
BackgroundFlexible Neuromorphic Devices Based on Two-Dimensional Transition Metal Dichalcogenides
Ma, X., Ding, G., Niu, W., Jia, Z., Han, S., Kuo, C., & Zhou, Y. (2023) · IEEE Journal on Flexible Electronics · doi:10.1109/jflex.2023.3298593 · Cited by 11
“Taking the image recognition task as an example, the high recognition accuracy requires a large dynamic range (e.g., the ratio of devices’ maximum and minimum conductivities) to enable excellent weight update (especially linearity and symmetry) [67], [71], [72].”
MethodUse of Magnetoresistive Random-Access Memory as Approximate Memory for Training Neural Networks
Locatelli, N., Vincent, A. F., & Querlioz, D. (2018) · IEEE International Conference on Electronics, Circuits and Systems (ICECS) · doi:10.1109/icecs.2018.8617952 · Cited by 11
Fast Object Tracking on a Many-Core Neural Network Chip
Deng, L., Zou, Z., Ma, X., Liang, L., Wang, G., Hu, X., Liu, L., Pei, J., Li, G., & Xie, Y. (2018) · Frontiers in Neuroscience · doi:10.3389/fnins.2018.00841 · Cited by 11
“The emerging non-volatile memory (NVM) devices (Yu, 2018) have demonstrated great potential to build ideal FunC with efficient processing of the mentioned vector/matrix operations, via integrating both computation and memory on the same memory crossbar in analog domain (Chi et al., 2016; Shafiee et…”
BackgroundCurrent‐Induced Planar Resistive Switching Mediated by Oxygen Migration in NiO/Pt Bilayer
Hu, S., Yang, H., Tang, M. X., Chen, H., Yang, Y., Zhou, S., & Qiu, X. (2020) · Advanced Electronic Materials · doi:10.1002/aelm.202000584 · Cited by 11
Rehman, S., Khan, M. F., Kim, H., & Kim, S. (2023) · Nano Energy · doi:10.1016/j.nanoen.2023.108333 · Cited by 10
Efficient Hybrid Training Method for Neuromorphic Hardware Using Analog Nonvolatile Memory
Kwon, D., Woo, S. Y., Hwang, J., Kim, H., Bae, J., Shin, W., Park, B., & Lee, J. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3327906 · Cited by 10
“While ANNs have achieved great success in various fields, state-of-the-art algorithm techniques in ANNs have been advancing toward increasing the network size, which vastly increases the computational complexity in the training and inference processes of networks [15].”
Recent Progress of Oxide Semiconductor-Based Artificial Synaptic Transistors
Lee, S. J., Park, J.-M., Ham, W., Lee, J., Hwang, G.-H., Namgung, S. D., Song, M., & Kwon, J. (2026) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.5c01798 · Cited by 10
Lee, C., Lee, J.-E., Kim, M.-J., Song, Y., Han, G., Seo, J., Kim, D., Seo, Y.-H., Hwang, H., & Lee, D. (2020) · Japanese Journal of Applied Physics · doi:10.35848/1347-4065/abd70d · Cited by 10
Efficient Low Cost Alternative Testing of Analog Crossbar Arrays for Deep Neural Networks
Ma, K., Saha, A., Amarnath, C., & Chatterjee, A. (2022) · IEEE International Test Conference (ITC) · doi:10.1109/itc50671.2022.00060 · Cited by 10
“However, RRAM crossbar-based neural networks [1] suffer from parametric manufacturing process variability effects resulting in as much as 30% degradation in performance (e.g., classification accuracy) [2], [3].”
BackgroundChen, Y.-F., Wang, C.-H., Shih, H.-Y., Kuo, C.-Y., & Wu, Y. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3374245 · Cited by 10
“The requirements of memory for IMC applications include: 1) high multilevel weights with high bit precision for synaptic operation; 2) high endurance for online training (repeated weight update); and 3) low power consumption for a sustainable environment [1].”
BackgroundYu, J., Kim, Y., Lee, C., Jeong, B., Kim, J., Han, J., Yang, J., Yun, S., Im, S. G., & Choi, Y. (2024) · Small · doi:10.1002/smll.202312283 · Cited by 11
Self-rectifying TiOx-based memristor with synaptic plasticity
Ye, Z., Wu, M., Xu, P., Chen, Z., Liang, X., Li, L., Meng, Y., Yang, H., & Ma, L. (2024) · Journal of Materials Science Materials in Electronics · doi:10.1007/s10854-024-12929-2 · Cited by 10
Feng, J., Liao, J., Jiang, Y., Bai, F., Zhu, J., Tang, X., Tang, Z., & Zhou, Y. (2024) · Materials Today Communications · doi:10.1016/j.mtcomm.2024.109805 · Cited by 10
Model of the Weak Reset Process in HfOx Resistive Memory for Deep Learning Frameworks
Majumdar, A., Bocquet, M., Hirtzlin, T., Laborieux, A., Klein, J., Nowak, E., Vianello, E., Portal, J., & Querlioz, D. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3108479 · Cited by 10
Glukhov, A., Milo, V., Baroni, A., Lepri, N., Zambelli, C., Olivo, P., Pérez, E., Wenger, C., & Ielmini, D. (2022) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps48227.2022.9764497 · Cited by 10
Cha, J., Jang, B. C., Oh, J., Lee, C., Yang, S. Y., Park, H., Im, S. G., & Choi, S. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200018 · Cited by 10
Lee, S., Woo, S. Y., & Lee, J. (2020) · IEEE Access · doi:10.1109/access.2020.3018226 · Cited by 10
An On-Chip Learning Method for Neuromorphic Systems Based on Non-Ideal Synapse Devices
Lee, J.-E., Lee, C., Kim, D.-W., Lee, D., & Seo, Y.-H. (2020) · Electronics · doi:10.3390/electronics9111946 · Cited by 10
“However, bottlenecks are often caused by the strict hardware requirements, excessive memory access, and high power consumption [3,4].”
BackgroundTotal Ionizing Dose Effects on Multistate HfOₓ-Based RRAM Synaptic Array
Han, X., Privat, A., Holbert, K. E., Seo, J.-S., Yu, S., & Barnaby, H. (2021) · IEEE Transactions on Nuclear Science · doi:10.1109/tns.2021.3072285 · Cited by 10
“the SRAM array, current leakage and limit to parallelism are major drawbacks to SRAM-based architectures [4].”
BackgroundInfluentialSelf-citationPal, P., Singh, A., & Wang, Y. (2023) · Journal of Physics D Applied Physics · doi:10.1088/1361-6463/acc8e4 · Cited by 9
Lateral Migration‐based Flash‐like Synaptic Device for Hybrid Off‐chip/On‐chip Training
Park, M., Hwang, J., Lee, K. M., Woo, S. Y., Kim, J., Bae, J., & Lee, J. (2024) · Advanced Electronic Materials · doi:10.1002/aelm.202300866 · Cited by 9
Youn, S., Kim, K., Park, J., & Kim, H. (2025) · Nano Energy · doi:10.1016/j.nanoen.2025.111633 · Cited by 9
Seong, S., Ha, T. T., Jung, S., Jeong, W., & Chung, Y. (2025) · Advanced Electronic Materials · doi:10.1002/aelm.202500521 · Cited by 10
Ag/HfO2-based Threshold Switching Memristor as an Oscillatory Neuron
Hua, Q., Jiang, C., & Hu, W. (2021) · IEEE Electron Devices Technology & Manufacturing Conference (EDTM) · doi:10.1109/edtm50988.2021.9420814 · Cited by 9
“Introduction Due to massive parallelism, fault-tolerance, and high energy-efficiency, neuromorphic systems composed of artificial neurons and artificial synapses show the promising potential to break the bottleneck of von Neumann architecture [1].”
BackgroundKim, H., Hwang, J., Kwon, D., Kim, J., Park, M., Im, J., Park, B., & Lee, J. (2021) · Advanced Intelligent Systems · doi:10.1002/aisy.202100064 · Cited by 9
Cao, T., Zhang, Z., Goh, W. L., Liu, C., Zhu, Y. H., & Gao, Y. (2023) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas57966.2023.10168639 · Cited by 9
“The performance of this work is compared with other recently reported CIM solutions for speech recognition in Tab.”
BackgroundInfluentialTotal Ionizing Dose Response of 128 Analog States in Computational Charge-Trap Memory
Xiao, T. P., Wilson, D., Bennett, C. H., Hughart, D. R., Feinberg, B., Agrawal, V. S., Puchner, H., Marinella, M. J., & Agarwal, S. (2023) · IEEE Transactions on Nuclear Science · doi:10.1109/tns.2023.3333812 · Cited by 9
“Analog IMC can address this need by using circuit laws to rapidly and efficiently compute a matrix–vector multiplication (MVM) [4], [5], [6], which is the dominant kernel in neural networks and many other algorithms.”
MethodProbabilistic Compute-in-Memory Design for Efficient Markov Chain Monte Carlo Sampling
Fu, Y., Shi, D., Fan, A., Yue, W., Yang, Y., Huang, R., & Yan, B. (2023) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2023.3337529 · Cited by 9
“As one of the promising solutions to overcome von Neumann bottleneck [16], [17], [18], [19], [20], this technology has been developed to perform energy-efficient computations in many AI applications [21], [22], [23], [24], [25], [26], [27], [28].”
MethodJeon, Y., Kim, J. H., Akinwande, D., & Choi, C. (2024) · Advanced Intelligent Systems · doi:10.1002/aisy.202300754 · Cited by 9
The Ouroboros of Memristors: Neural Networks Facilitating Memristor Programming
Yu, Z., Yang, M.-J., Finkbeiner, J., Siegel, S., Strachan, J. P., & Neftci, E. (2024) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas59952.2024.10595913 · Cited by 10
“The noise-free model faces challenges near G min and G max due to the self-accelerating processes discussed in Sec.”
BackgroundOptical Bio-Inspired Synaptic Devices
Li, P., Wang, K., Jiang, S., He, G., Zhang, H., Cheng, S., Li, Q., Zhu, Y., Fu, C., Wei, H., et al. (2024) · Nanomaterials · doi:10.3390/nano14191573 · Cited by 9
A Construction Kit for Efficient Low Power Neural Network Accelerator Designs
Jokic, P., Azarkhish, E., Bonetti, A., Pons, M., Emery, S., & Benini, L. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3520127 · Cited by 11
“Their previous survey [67] provides a comprehensive list of the utilized emerging NVMs, illustrating that NVM can increase bit density and reduce leakage compared to SRAM and possibly store multiple bits per cell.”
BackgroundIon beam-assisted solid phase epitaxy of SiGe and its application for analog memristors
Kim, K.-H., Kang, D. C., Jeong, Y., Kim, J., Lee, S., Kwak, J. Y., Park, J., Hwang, G. W., Lee, K.-S., Ju, B., et al. (2021) · Journal of Alloys and Compounds · doi:10.1016/j.jallcom.2021.161086 · Cited by 10
NVM Device-Based Deep Inference Architecture Using Self-gated Activation Functions (Swish)
Fatima, A., & Pethe, A. J. (2021) · Lecture notes in electrical engineering · doi:10.1007/978-981-16-5078-9_4 · Cited by 9
Understanding composite negative differential resistance in niobium oxide memristors
Liu, X., Zhang, P., Nath, S. K., Li, S., Nandi, S. K., & Elliman, R. G. (2021) · Journal of Physics D Applied Physics · doi:10.1088/1361-6463/ac3bf4 · Cited by 9
Mnemonic-opto-synaptic transistor for in-sensor vision system
Han, J., Chung, Y., Sim, J., Yu, J., Lee, G., Kim, S., & Choi, Y. (2022) · Scientific Reports · doi:10.1038/s41598-022-05944-y · Cited by 9
Kim, S., Yu, J., Lee, G. S., Yun, D.-H., Kim, M., Kim, J., Kim, D., Lee, G., Kim, M., Han, J., et al. (2022) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.2c07975 · Cited by 9
Mechanism of the Resistivity Switching Induced by the Joule Heating in Crystalline NbO 2
Olin, S., Razek, S. A., Piper, L. F. J., & Lee, W. (2022) · Advanced Quantum Technologies · doi:10.1002/qute.202200067 · Cited by 9
Demonstration of Pulse Width Modulation Function Using Single Positive Feedback Device for Neuron
Woo, S. Y., Kwon, D., Park, B., Lee, J., & Bae, J. (2022) · IEEE Electron Device Letters · doi:10.1109/led.2022.3225168 · Cited by 9
“R ECENTLY, deep neural networks (DNNs) are attracting attention with inference performance comparable to or higher than humans for specific tasks, such as recognizing hand-written digits and classifying certain images [1], [2], [3].”
BackgroundPyle, S. D., Zand, R., Sheikhfaal, S., & DeMara, R. F. (2019) · IEEE Journal on Exploratory Solid-State Computational Devices and Circuits · doi:10.1109/jxcdc.2019.2911046 · Cited by 11
“R ECENT research into spiking neural network (SNN) hardware has aimed to achieve computational capabilities akin to biological brains, such as inference, at efficien-cies unachievable with Von-Neumann hardware [1].”
BackgroundEffect of Asymmetric Nonlinearity Dynamics in RRAMs on Spiking Neural Network Performance
Fouda, M. E., Neftci, E., Eltawil, A. M., & Kurdahi, F. J. (2019) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/ieeeconf44664.2019.9049043 · Cited by 9
“Several RRAM devices demonstrating promising synaptic behaviors are characterized by nonlinear and asymmetric update dynamics, which is a major obstacle for large-scale deployment in neural networks [8], especially for learning tasks.”
BackgroundROA: A Rapid Learning Scheme for In-Situ Memristor Networks
Zhang, W., Wang, Y., Ji, X., Wu, Y., & Zhao, R. J. (2021) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2021.692065 · Cited by 9
“Implementation of neural networks in memristor-based hardware exhibits high density integration, low power consumption and high ef fi ciency (Cai et al., 2019; Yu 2018).”
BackgroundSimanjuntak, F. M., Hsu, C.-L., Abbey, T., Chang, L.-Y., Rajasekaran, S., Prodromakis, T., & Tseng, T. (2021) · APL Materials · doi:10.1063/5.0067302 · Cited by 9
Kim, M., Kim, J., Yun, G., Yu, J., Han, J., Lee, J.-W., Seo, S., Choi, S., & Choi, Y. (2022) · IEEE Electron Device Letters · doi:10.1109/led.2022.3194556 · Cited by 9
“every synapse is updated through each training cycle [9].”
BackgroundJiang, Y., Li, L., Wang, J., Yin, M., Liu, H., Dai, Z., Cui, X., & Cui, X. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10558351 · Cited by 8
“However, implementation of SNN training algorithms on hardware remains a challenge [6], [7], [10]–[13].”
BackgroundAI-driven model for optimized pulse programming of memristive devices
Spetzler, B., Fritscher, M., Park, S., Kim, N., Wenger, C., & Ziegler, M. (2025) · APL Machine Learning · doi:10.1063/5.0251113 · Cited by 9
Enhancing RRAM Reliability: Exploring the Effects of Al Doping on HfO 2 -Based Devices
Baroni, A., Pérez, E., Reddy, K. D. S., Pechmann, S., Wenger, C., Ielmini, D., & Zambelli, C. (2025) · IEEE Transactions on Device and Materials Reliability · doi:10.1109/tdmr.2025.3581061 · Cited by 8
“RRAM arrays enable energy-efficient and high-performance designs for Vector-Matrix Multiplication (VMM), a key operation in Artificial Neural Networks (ANN) [5], [6], [7], [8].”
A new era of ferroelectric thin films for nonvolatile memories
Trolier-McKinstry, S. E., Uchida, K., & Yamada, T. (2025) · MRS Bulletin · doi:10.1557/s43577-025-00969-w · Cited by 8
An overhead-reduced, efficient, fully analog neural-network computing hardware
Ye, J., Wang, W., Shi, C., Cui, X., Qu, W., Lin, P., Yu, X., Cheng, R., Wu, H., & Chen, B. (2025) · Science Advances · doi:10.1126/sciadv.adv7555 · Cited by 10
Memristor Synapse—A Device-Level Critical Review
Chandrasekaran, S., Chang, Y., & Simanjuntak, F. M. (2026) · Nanomaterials · doi:10.3390/nano16030179 · Cited by 9
“Nevertheless, the current computer architecture is based on conventional von Neumann designs that rely on data transfer between the processing unit and memory unit, and its transfer rate is facing a limit as the task complexity increases [2].”
InfluentialNirmal, K. A., Patil, A. R., Patil, P. P., Khot, A. C., Dongale, T. D., & Kim, T. G. (2026) · International Journal of Extreme Manufacturing · doi:10.1088/2631-7990/ae542d · Cited by 8
“In artificial synapses, the DR is more critical than the ON/OFF ratio because it directly affects precision and learning efficiency [77,78] .”
Neuromorphic Computing between Reality and Future Needs
Ahmed, K. S., & Shereif, F. F. (2023) · Artificial intelligence · doi:10.5772/intechopen.110097 · Cited by 10
Sun, J. W., Wang, Z., Wang, S., Cao, Y., Gao, H., Wang, H., Ma, X., & Hao, Y. (2023) · Applied Physics Letters · doi:10.1063/5.0146341 · Cited by 8
Zhou, X., Wu, W., Gu, H., Zhang, P., Fu, B., Zhu, X., & Zhai, J. (2023) · Ceramics International · doi:10.1016/j.ceramint.2023.10.132 · Cited by 8
Non von Neumann computing concepts
Syed, G. S., Gallo, M. L., & Sebastian, A. (2024) · Elsevier eBooks · doi:10.1016/b978-0-12-823491-4.00002-3 · Cited by 10
Zhang, J., Ren, X., Pan, Y., Nan, Y., & Zheng, H. (2024) · Ceramics International · doi:10.1016/j.ceramint.2024.04.156 · Cited by 8
Kim, D., Truong, P. L., Lee, C. B., Bang, H., Choi, J., Ham, S. Y., Ko, J. H., Kim, K., Lee, D., & Park, H. J. (2024) · Small · doi:10.1002/smll.202402961 · Cited by 10
Passive LiNbO₃ Memristor With Multilevel States for Neuromorphic Computing
Xie, Q., Pan, X., Wang, Y., Luo, W., Zhao, Z., Tong, J., Yang, X., Shuai, Y., Wu, C., & Zhang, W. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3450437 · Cited by 8
“Memristors, as the most promising device for neuromorphic computing [6], [7], [8], [9].”
BackgroundUm, M., Kang, M., Eom, K., Kwak, H., Noh, K., Lee, J., Son, J., Kwon, J., Kim, S., & Lee, H. (2024) · IEEE Transactions on Biomedical Circuits and Systems · doi:10.1109/tbcas.2024.3465610 · Cited by 9
“When operating on a graphics processing unit (GPU), there is intensive data exchange between the processor and memory, leading to signi fi cant energy consumption [4].”
Probabilistic metaplasticity for continual learning with memristors in spiking networks
Zohora, F. T., Karia, V., Soures, N., & Kudithipudi, D. (2024) · Scientific Reports · doi:10.1038/s41598-024-78290-w · Cited by 8
“Hence, edge devices should learn in an online scenario where data are encountered only once [2].”
Jung, G., Kim, H., Jeong, Y., Hong, Y., Wu, M., Hong, S., Shin, W., Jang, D., & Lee, J.-H. (2019) · IEEE International Symposium on Olfaction and Electronic Nose (ISOEN) · doi:10.1109/isoen.2019.8823181 · Cited by 8
“A neuromorphic computing technology using nonvolatile memory devices have been studied [11], and it is meaningful to implement the electronic nose by applying synaptic imitation devices with nonvolatile memory function to DNN.”
MethodA SPICE compact model for forming-free, low-power graphene-insulator-graphene ReRAM technology
Reddy, L. H., Pande, S. R., Roy, T., Vogel, E. M., Chakravorty, A., & Chakrabarti, B. (2021) · Emergent Materials · doi:10.1007/s42247-021-00265-8 · Cited by 8
“ReRAMs with sub-ns switching speed and non-volatility can be useful as a universal memory, bridging the gap between fast but volatile DRAMs and slower but non-volatile flash [7].”
BackgroundTackling the Low Conductance State Drift through Incremental Reset and Verify in RRAM arrays
Baroni, A., Zambelli, C., Olivo, P., Pérez, E., Wenger, C., & Ielmini, D. (2021) · IEEE International Integrated Reliability Workshop final report · doi:10.1109/iirw53245.2021.9635613 · Cited by 8
Synaptic transistors based on a tyrosine-rich peptide for neuromorphic computing
Song, M., Song, Y., Sung, T., Namgung, S. D., Yoon, J. H., Lee, Y., Nam, K. T., & Kwon, J. (2021) · RSC Advances · doi:10.1039/d1ra06492d · Cited by 8
Duan, H., Li, E., Yang, Y., & Chen, W. (2022) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2022.3206724 · Cited by 8
“readout, fast write/read speed, and low power consumption [1], [2], [3].”
BackgroundArtificial synapse based on 1,4-diphenylbutadiyne with femtojoule energy consumption
Liu, J., Li, Z., Jia, C., & Zhang, W. (2023) · Physical Chemistry Chemical Physics · doi:10.1039/d2cp05417e · Cited by 8
A Hybrid Weight Quantization Strategy for Memristive Neural Networks
Shen, S., Duan, S., & Wang, L. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126778 · Cited by 9
Jeon, Y., Kim, D., Ku, B., Chung, C., & Choi, C. (2023) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.3c13159 · Cited by 12
Electrochemical memristive devices based on submonolayer metal deposition
Pragnya, P., Pinkowitz, A., Hull, R. V., & Gall, D. (2019) · APL Materials · doi:10.1063/1.5110889 · Cited by 8
Jang, J. T., Min, J., Kim, D., Park, J., Choi, S., Kim, D. M., Cho, S., & Kim, D. H. (2020) · Solid-State Electronics · doi:10.1016/j.sse.2020.107764 · Cited by 8
Bidirectional Analog Conductance Modulation for RRAM-Based Neural Networks
Jiang, Z., Wang, Z., Zheng, X., Fong, S. W., Qin, S., Chen, H.-Y., Ahn, C., Cao, J., Nishi, Y., Wong, S. S., et al. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.3025849 · Cited by 9
“One of the key challenges for neural networks employing RRAM is to achieve bidirectional analog conductance modulation of RRAM for online training [7], [8].”
Xiong, W., Liu, L., Liu, J., Chen, W., & Zheng, Y. (2021) · Applied Physics Letters · doi:10.1063/5.0050847 · Cited by 8
SCANet: Securing the Weights With Superparamagnetic-MTJ Crossbar Array Networks
Rajasekharan, D., Rangarajan, N., Patnaik, S., Sinanoglu, O., & Chauhan, Y. S. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3130884 · Cited by 8
“by reverse-engineering these weights that are represented by the programmable conductance of nonvolatile memories (NVM), such as resistive random access memories (RRAM), phase change memories (PCM), and magnetic tunnel junctions (MTJs) [21].”
Xu, Z., Bernussi, A. A., & Fan, Z. (2022) · Electronics · doi:10.3390/electronics11040516 · Cited by 8
Hardware/software Co-design for Neuromorphic Systems
Manohar, R. (2022) · Custom Integrated Circuits Conference · doi:10.1109/cicc53496.2022.9772863 · Cited by 8
Memristor-based Deep Convolution Neural Network: A Case StudyPreprint
Zhang, F., & Hu, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.02225 · Cited by 8
“eneck when dealing with high volume convolution calculations[7], [8]. Recently, many researchers show interest on emerging memristor crossbar for its computing-in-memory feature[9], [10], [11], [12], [13], [14]. A memristor crossbar can carry out large-scale analog vector matrix multiplication in one step by collectin…”
Mishra, A. K., Pathak, A., & Pandey, S. K. (2025) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.4c01802 · Cited by 7
Bhardwaj, K., Paasio, E., & Majumdar, S. (2025) · Advanced Intelligent Discovery · doi:10.1002/aidi.202500143 · Cited by 7
Kang, H. Y., Shin, Y. H., Kim, D. E., Kwon, D. W., & Jeong, J. K. (2026) · Nanoscale Advances · doi:10.1039/d5na00980d · Cited by 7
Demonstration of integrate-and-fire neuron circuit for spiking neural networks
Woo, S. Y., Kang, W.-M., Seo, Y.-T., Lee, S., Kwon, D., Oh, S., Bae, J., & Lee, J. (2022) · Solid-State Electronics · doi:10.1016/j.sse.2022.108481 · Cited by 7
Compute-in-Memory Architecture
Jiang, H., Huang, S., & Yu, S. (2023) · Springer eBooks · doi:10.1007/978-981-15-6401-7_62-1 · Cited by 7
Self-citationMeasuring Electrical Resistivity at the Nanoscale in Phase-Change Materials
Zhang, L., Lorut, F., Gruel, K., Hÿtch, M., & Gatel, C. (2024) · Nano Letters · doi:10.1021/acs.nanolett.4c01462 · Cited by 7
Kang, M., Cho, U., Kang, J., Han, N., Seo, H. J., Yang, J., Shin, S., Kim, T., Kim, S., Jeong, C., et al. (2024) · Advanced Electronic Materials · doi:10.1002/aelm.202300900 · Cited by 7
Lee, J. Y., Kim, C. J., Ku, M., Kim, T., Noh, T., Lee, S. W., Shin, Y. G., Ahn, J., & Kang, B. S. (2024) · Current Applied Physics · doi:10.1016/j.cap.2024.07.018 · Cited by 7
Nano Horizons: Exploring the untapped power of two-Dimensional materials
Kim, J., Kumar, A., Bhoyate, S. D., Hwang, J., Jang, H., Mahajan, C., Lee, E., & Gupta, R. K. (2024) · Materials Science and Engineering B · doi:10.1016/j.mseb.2024.117673 · Cited by 8
Van der Waals Inverted-Floating-Gate Transistors for Artificial Intelligence Electronics
Soliman, M., Marchand, C., Mahmoudi, A., Rajak, N. K., Taniguchi, T., Watanabe, K., Gloppe, A., Doudin, B., Deleruyelle, D., O’Connor, I., et al. (2025) · ACS Nano · doi:10.1021/acsnano.5c03875 · Cited by 7
Metadata
- Authors
- Yu, S.
- Affiliations
- Arizona State University (US)
- Corresponding author
- Yu, Shimeng
- ORCID
- Yu, Shimeng
- Venue
- Proceedings of the IEEE
- Published
- 2018-02
- Volume
- 106
- Issue
- 2
- Pages
- 260-285
- ISSN
- 0018-9219 (ISSN-L)
- 1558-2256
- OpenAlex
- W2785141883
- Cited by
- 1,170
- References
- 141
- Citations by year
- 2026: 96 · 2025: 139 · 2024: 138 · 2023: 148 · 2022: 177 · 2021: 146 · 2020: 162 · 2019: 117 · 2018: 47
- FWCI
- 45.15
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- Proceedings of the IEEE (publishedVersion)
- Funders
- NSF-CCF (NSF-CCF-1552687)
- NSFCCF (NSFCCF-1715443)
- NSF-ECCS (NSF-ECCS-1701565)
- HDTRA (HDTRA1-17-1-0038)
- NSF/SRC (NSF/SRC E2CDA)
- ASCENT (one of the six SRC/DARPA JUMP Centers)
- Qualcomm
- Samsung
- 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 Reservoir Computing
- Keywords
- neuromorphic computing, neuromorphic hardware, non-volatile memory, artificial synapses, phase-change memory, resistive random access memory, ferroelectric memory, floating-gate transistors, pattern recognition
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.