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 201–300 of 1,001
Recent Research for HZO-Based Ferroelectric Memory towards In-Memory Computing Applications
Yoo, J., Song, H., Lee, H., Lim, S., Kim, S., Heo, K., & Bae, H. (2023) · Electronics · doi:10.3390/electronics12102297 · Cited by 61
Yu, E., Gaurav Kumar, K., Saxena, U., & Roy, K. (2024) · Scientific Reports · doi:10.1038/s41598-024-59298-8 · Cited by 63
Jebali, F., Majumdar, A., Turck, C., Harabi, K.-E., Faye, M.-C., Muhr, E., Walder, J.-P., Bilousov, O., Michaud, A., Vianello, E., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-44766-6 · Cited by 58
“A promising solution to this problem is the use of memristor-based systems, which can drastically reduce the energy consumption of AI 5,6 , making it even conceivable to create self-powered edge AI systems that do not require batteries and can instead harvest energy from the environment.”
A memristor-based Bayesian machine
Harabi, K.-E., Hirtzlin, T., Turck, C., Vianello, E., Laurent, R., Droulez, J., Bessière, P., Portal, J.-M., Bocquet, M., & Querlioz, D. (2022) · Nature Electronics · doi:10.1038/s41928-022-00886-9 · Cited by 58
Nandi, S. K., Nath, S. K., El‐Helou, A. E., Li, S., Liu, X., Raad, P. E., & Elliman, R. G. (2019) · Advanced Functional Materials · doi:10.1002/adfm.201906731 · Cited by 58
Neuromorphic Computing Using Emerging Synaptic Devices: A Retrospective Summary and an Outlook
Park, J. (2020) · Electronics · doi:10.3390/electronics9091414 · Cited by 58
Kundu, S., Ganganaik, P. B., Louis, J., Chalamalasetty, H., & Rao, B. P. (2022) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2022.3161847 · Cited by 57
2D-Material-Based Volatile and Nonvolatile Memristive Devices for Neuromorphic Computing
Xia, X., Huang, W., Hang, P., Guo, T., Yan, Y., Yang, J., Yang, D., Yu, X., & Li, X. (2023) · ACS Materials Letters · doi:10.1021/acsmaterialslett.2c01026 · Cited by 59
Neuromorphic Time-Dependent Pattern Classification with Organic Electrochemical Transistor Arrays
Pecqueur, S., Talamo, M. M., Guérin, D., Blanchard, P., Roncali, J., Vuillaume, D., & Alibart, F. (2018) · LillOA (Université de Lille (University Of Lille)) · Cited by 57
Chen, Y., Li, D., Ren, H., Tang, Y., Liang, K., Wang, Y., Li, F., Song, C., Guan, J., Chen, Z., et al. (2022) · Small · doi:10.1002/smll.202203611 · Cited by 56
A 2D material-based floating gate device with linear synaptic weight update
Park, E., Kim, M., Kim, T. S., Kim, I. S., Park, J., Kim, J., Jeong, Y., Lee, S., Kim, I., Park, J., et al. (2020) · Nanoscale · doi:10.1039/d0nr07403a · Cited by 56
Variability and Energy Consumption Tradeoffs in Multilevel Programming of RRAM Arrays
Pérez, E., Mahadevaiah, M. K., Quesada, E. P.-B., & Wenger, C. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3072868 · Cited by 59
“RESISTIVE random access memory (RRAM) technology based on metal-oxide dielectrics is one of the most promising candidates for implementing the next generation of nonvolatile memories (NVMs) as well as the synaptic weights integrated into hardware-based artificial neural networks (ANNs) [1], [2].”
BackgroundStuck-at-Fault Tolerant Schemes for Memristor Crossbar Array-Based Neural Networks
Yeo, I., Chu, M., Gi, S.-G., Hwang, H., & Lee, B.-G. (2019) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2019.2914460 · Cited by 55
“VECTOR–MATRIX multiplication (VMM), which requires high-density memories, is one of the most common but energy-expensive computational operations in neural networks (NNs) [1]–[3].”
BackgroundDesign Considerations for Efficient Deep Neural Networks on Processing-in-Memory Accelerators
Yang, T.-J., & Sze, V. (2019) · 2019 IEEE International Electron Devices Meeting (IEDM) · doi:10.1109/iedm19573.2019.8993662 · Cited by 55
“While there are existing efforts to explore the use of PIM accelerators for running DNNs, most of them target smallscale DNNs – typically 2 or 3 layer multi-layer perceptrons – for the simple problem of digit classification on the MNIST dataset [2].”
BackgroundMethodInfluentialOh, S., Lee, J., Seo, S., Choo, H., Lee, D., Cho, J., & Park, J. (2021) · Advanced Science · doi:10.1002/advs.202103808 · Cited by 60
Luo, X., Ming, J., Gao, J., Zhuang, J., Fu, J., Ren, Z., Ling, H., & Xie, L. (2022) · Frontiers in Neuroscience · doi:10.3389/fnins.2022.1016026 · Cited by 58
“The small non-linearity (∼1) can effectively reduce the time redundancy and energy loss in the training process of neural networks (Yu, 2018; Yang F. et al., 2021).”
BackgroundRajput, M., Mallik, S. K., Chatterjee, S., Shukla, A., Hwang, S., Sahoo, S., Kumar, G. V. P., & Rahman, A. (2024) · Communications Materials · doi:10.1038/s43246-024-00632-y · Cited by 59
“A higher number of multilevel states (e.g., exceeding hundreds of levels) contribute to improved learning capability and enhance network robustness 82 .”
Two-step write–verify scheme and impact of the read noise in multilevel RRAM-based inference engine
Shim, W., Seo, J.-S., & Yu, S. (2020) · Semiconductor Science and Technology · doi:10.1088/1361-6641/abb842 · Cited by 61
Self-citationA self-adaptive hardware with resistive switching synapses for experience-based neurocomputing
Bianchi, S., Muñoz-Martín, I., Covi, E., Bricalli, A., Piccolboni, G., Regev, A., Molas, G., Nodin, J. F., Andrieu, F., & Ielmini, D. (2023) · Nature Communications · doi:10.1038/s41467-023-37097-5 · Cited by 58
So, H., Lee, J., Mahata, C., Kim, S., & Kim, S. (2024) · Advanced Materials Technologies · doi:10.1002/admt.202301390 · Cited by 54
Impact of RTN on Pattern Recognition Accuracy of RRAM-Based Synaptic Neural Network
Chai, Z., Freitas, P., Zhang, W., Hatem, F. O., Zhang, J. F., Marsland, J., Govoreanu, B., Goux, L., & Kar, G. S. (2018) · IEEE Electron Device Letters · doi:10.1109/led.2018.2869072 · Cited by 54
In-Memory PageRank Accelerator With a Cross-Point Array of Resistive Memories
Sun, Z., Ambrosi, E., Pedretti, G., Bricalli, A., & Ielmini, D. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.2966908 · Cited by 52
“The mean value μG of the conductance shows an almost linear increase from 2 to 32 μS, which supports our RRAM device for implementing analog matrix entries [13].”
MethodWu, T.-Y., Huang, H., Chu, Y.-H., Chang, C.-C., Wu, M.-H., Hsu, C.-H., Wu, C.-T., Wu, M.-C., Wu, W., Chang, T., et al. (2019) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm19573.2019.8993565 · Cited by 54
“Introduction By exploiting the highly parallel vector-matrix multiplication (VMM) in local memory arrays without data transfer, recently in-memory computing (IMC) is widely researched for accelerating deep learning [1, 2].”
BackgroundChen, Y., Lin, C., & Chang, Y. (2021) · Micromachines · doi:10.3390/mi12010050 · Cited by 54
“…is also a promising candidate for next-generation nonvolatile memory technologies for both data storage and in-memory computing owing to its ultra-low power consumption, simple structure, and high scalability for 3D integration of the computing and memory elements, e.g. 3D vertical RRAM [11–14].”
BackgroundFerroelectric memory based on two-dimensional materials for neuromorphic computing
Chen, L., Pam, M. E., Li, S., & Ang, K. (2022) · Neuromorphic Computing and Engineering · doi:10.1088/2634-4386/ac57cb · Cited by 55
In-memory computing with ferroelectrics
Yang, R. (2020) · Nature Electronics · doi:10.1038/s41928-020-0411-2 · Cited by 56
Two‐Dimensional MXene Synapse for Brain‐Inspired Neuromorphic Computing
Ju, J. H., Seo, S., Baek, S., Lee, D., Lee, S., Lee, T., Kim, B., Lee, J., Koo, J., Choo, H., et al. (2021) · Small · doi:10.1002/smll.202102595 · Cited by 53
Electrolyte-gated neuromorphic transistors for brain-like dynamic computing
He, Y., Jiang, S., Chen, C., Wan, C., Shi, Y., & Wan, Q. C. (2021) · Journal of Applied Physics · doi:10.1063/5.0069456 · Cited by 55
Stochastic Computing for Hardware Implementation of Binarized Neural Networks
Hirtzlin, T., Penkovsky, B., Bocquet, M., Klein, J., Portal, J., & Querlioz, D. (2019) · IEEE Access · doi:10.1109/access.2019.2921104 · Cited by 57
Shen, J., Zhou, B., Wang, F., Wan, Q. C., Shan, X., Li, C., Lin, X., & Zhang, K. (2020) · Nanotechnology · doi:10.1088/1361-6528/ab82d6 · Cited by 53
Kwon, D., Jung, G., Shin, W., Jeong, Y., Hong, S., Oh, S., Kim, J., Bae, J., Park, B., & Lee, J.-H. (2021) · Sensors and Actuators B Chemical · doi:10.1016/j.snb.2021.130419 · Cited by 57
Khan, A. I., Yu, H., Zhang, H. R., Goggin, J. R., Kwon, H., Wu, X., Perez, C., Neilson, K. M., Asheghi, M., Goodson, K. E., et al. (2023) · Advanced Materials · doi:10.1002/adma.202300107 · Cited by 52
Xiang, Y., Huang, P. C., Zhao, Y., Zhao, M., Gao, B., Wu, H., Qian, H., Liu, X., & Kang, J. (2019) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2019.2931135 · Cited by 56
Chen, Y., Lin, C., Hu, S.-T., Lin, C., Fowler, B. W., & Lee, J. (2019) · Scientific Reports · doi:10.1038/s41598-019-48932-5 · Cited by 53
Shen, P., Lin, C., Wang, H., Teo, K. H., & Kong, J. (2020) · Applied Physics Letters · doi:10.1063/1.5129963 · Cited by 52
Mulaosmanovic, H., Dünkel, S., Trentzsch, M., Beyer, S., Breyer, E. T., Mikolajick, T., & Slesazeck, S. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.3031249 · Cited by 57
“, electrical disturbs coming from write and read operations in FeFET memory arrays are generally in the form of a train of voltage pulses; artificial synapses in deep neural networks tend to be potentiated and depressed with identical pulses [22], [23]; and the accumulation of pulses and the subsequent FeFET switching…”
BackgroundA Multi-functional Memristive Pavlov Associative Memory Circuit Based on Neural Mechanisms
Zhang, Y., & Zeng, Z. (2021) · IEEE Transactions on Biomedical Circuits and Systems · doi:10.1109/tbcas.2021.3108354 · Cited by 54
Outstanding Bit Error Tolerance of Resistive RAM-Based Binarized Neural Networks
Hirtzlin, T., Bocquet, M., Klein, J.-O., Nowak, E., Vianello, E., Portal, J.-M., & Querlioz, D. (2019) · 2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas.2019.8771544 · Cited by 51
Simanjuntak, F. M., Ohno, T., Chandrasekaran, S., Tseng, T., & Samukawa, S. (2020) · Nanotechnology · doi:10.1088/1361-6528/ab7fcf · Cited by 53
“This arose from the high nonlinearity of the as-deposited device, which causes a training accuracy loss in neural networks.[1]”
BackgroundMethodInfluentialProspect and challenges of analog switching for neuromorphic hardware
Banerjee, W., Nikam, R. D., & Hwang, H. (2022) · Applied Physics Letters · doi:10.1063/5.0073528 · Cited by 51
Functional Oxides for Photoneuromorphic Engineering: Toward a Solar Brain
Pérez‐Tomás, A. (2019) · Advanced Materials Interfaces · doi:10.1002/admi.201900471 · Cited by 52
Zhou, Z., Zhou, J., Wang, X., Wang, H., Sun, C., Han, K., Kang, Y., Zheng, Z., Ni, H., & Gong, X. (2020) · IEEE Electron Device Letters · doi:10.1109/led.2020.3035276 · Cited by 51
“) pulse scheme, the linearity of the programming and erasing is comparable with those of other eNVMs [20], which indicates the potential of this device as a multi-level memory cell.”
ResultFully CMOS-compatible passive TiO2-based memristor crossbars for in-memory computing
Mesoudy, A. E., Lamri, G., Dawant, R., Arias‐Zapata, J., Gliech, P., Beilliard, Y., Ecoffey, S., Ruediger, A. P., Alibart, F., & Drouin, D. (2022) · Microelectronic Engineering · doi:10.1016/j.mee.2021.111706 · Cited by 51
Energy- and Area-Efficient CMOS Synapse and Neuron for Spiking Neural Networks With STDP Learning
Joo, B., Han, J., & Kong, B. (2022) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2022.3178989 · Cited by 55
“It is noted that the energy consumption of the proposed synapse is comparable to that of a biological synapse to be measured around 4 ∼ 50 fJ/spike [37], [38].”
MethodTechnological Benchmark of Analog Synaptic Devices for Neuroinspired Architectures
Chen, P.-Y., & Yu, S. (2018) · IEEE Design and Test · doi:10.1109/mdat.2018.2890229 · Cited by 52
“The most compact and simplest array structure to form a weight matrix is the crossbar array structure with eNVM analog synapses [15].”
BackgroundSelf-citationZhong, S., Su, L., Xu, M., Loke, D. K., Yu, B., Zhang, Y., & Zhao, R. J. (2024) · Nano-Micro Letters · doi:10.1007/s40820-024-01550-x · Cited by 53
“The realization of AI relies heavily on computing hardware, which is widely utilized with conventional von Neumann architecture [6, 7].”
Lee, J.-H., Lim, D., Jeong, H., Ma, H., & Shi, L. (2019) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2019.2906249 · Cited by 58
“Recently, many researchers have studied memristors, such as phase-change random access memory (PCRAM) [4].”
BackgroundWoo, J., & Yu, S. (2019) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2019.2917764 · Cited by 48
“The crossbar array, where word line (WL) and bit line (BL) are perpendicular to each other, could accelerate the neuromorphic computation in a two-stage process [13].”
BackgroundSelf-citationShim, W., Luo, Y., Seo, J.-S., & Yu, S. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.2985013 · Cited by 57
“have been investigated, such as resistive random access memory (RRAM) [1]–[6], phase change random access memory”
BackgroundSelf-citationWan, F., Wang, Q., Harumoto, T., Gao, T., Ando, K., Nakamura, Y., & Shi, J. F. (2020) · Advanced Functional Materials · doi:10.1002/adfm.202007101 · Cited by 51
Projected Mushroom Type Phase‐Change Memory
Sarwat, S. G., Philip, T. M., Chen, C.-T., Kersting, B., Bruce, R. L., Cheng, C., Li, N., Saulnier, N., BrightSky, M. J., & Sebastian, A. (2021) · Advanced Functional Materials · doi:10.1002/adfm.202106547 · Cited by 55
Reconfigurable Field-Effect Transistor as a Synaptic Device for XNOR Binary Neural Network
Bae, J., Kim, H., Kwon, D., Lim, S., Lee, S., Park, B., & Lee, J. (2019) · IEEE Electron Device Letters · doi:10.1109/led.2019.2898448 · Cited by 61
“geous for implementing both software-based and hardwarebased neural networks [5], [11].”
BackgroundMethodTao, J., Sarkar, D., Kale, S., Singh, P. K., & Kapadia, R. (2020) · Nano Letters · doi:10.1021/acs.nanolett.0c03548 · Cited by 49
Jeong, Y., Lee, H., Ryu, D. G., Cho, S. H., Lee, G., Kim, S., Kim, S., & Lee, Y. S. (2021) · Advanced Electronic Materials · doi:10.1002/aelm.202100185 · Cited by 49
Two-dimensional materials for artificial sensory devices: advancing neuromorphic sensing technology
Ko, J., Ock, C., Gim, H., Hong, K., Lee, Y., & Kwon, K. C. (2025) · npj 2D Materials and Applications · doi:10.1038/s41699-025-00556-2 · Cited by 51
A 40nm Low-Power Logic Compatible Phase Change Memory Technology
Wu, J. Y., Chen, Y. S., Khwa, W.-S., Yu, S. M., Wang, T. Y., Tseng, J. C. C., Chih, Y. D., & Diaz, C. H. (2018) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm.2018.8614513 · Cited by 48
Self-citationJang, J. T., Ahn, G., Choi, S., Kim, D. M., & Kim, D. H. (2019) · Electronics · doi:10.3390/electronics8101087 · Cited by 50
“In particular, the synapse device using the memristor requires excellent linearity according to the consecutive potentiation/depression pulse for high data processing accuracy [11].”
MethodCMOS-compatible synaptic transistor gated by chitosan electrolyte-Ta2O5 hybrid electric double layer
Min, S.-Y., & Cho, W.-J. (2020) · Scientific Reports · doi:10.1038/s41598-020-72684-2 · Cited by 48
Ferroelectric HfO 2 -based synaptic devices: recent trends and prospects
Yu, S., Hur, J. H., Luo, Y.-C., Shim, W., Choe, G., & Wang, P. (2021) · Semiconductor Science and Technology · doi:10.1088/1361-6641/ac1b11 · Cited by 52
Self-citationAtomically Thin Synapse Networks on Van Der Waals Photo‐Memtransistors
Moon, G., Min, S. Y., Han, C., Lee, S., Ahn, H., Seo, S., Ding, F., Kim, S., & Jo, M. (2022) · Advanced Materials · doi:10.1002/adma.202203481 · Cited by 48
Precision of bit slicing with in-memory computing based on analog phase-change memory crossbars
Le Gallo, M., Nandakumar, S. R., Ciric, L., Boybat, I., Khaddam-Aljameh, R., Mackin, C., & Sebastian, A. (2022) · Neuromorphic Computing and Engineering · doi:10.1088/2634-4386/ac4fb7 · Cited by 42
“In-memory computing cores typically consist of a memory array, data converters and a digital control unit [1, 2].”
BackgroundControlling sulfurization of 2D Mo2C crystal for Mo2C/MoS2-based memristor and artificial synapse
Tang, X., Yang, L., Huang, J., Chen, W., Li, B., Yang, S., Yang, R., Zeng, Z., Tang, Z., & Gui, X. (2022) · npj Flexible Electronics · doi:10.1038/s41528-022-00227-y · Cited by 47
On-chip direct synthesis of boron nitride memristors
Xie, J., Yekta, A. E., Mamun, F. A., Zhu, K., Chen, M., Pazos, S. M., Zheng, W., Zhang, X., Tongay, S., Li, X., et al. (2025) · Nature Nanotechnology · doi:10.1038/s41565-025-01988-z · Cited by 48
“https://doi.org/10.1038/s41565-025-01988-z of memristive devices is also important for accuracy 47,48 .”
InfluentialOh, S., Shi, Y., Liu, X., Song, J., & Kuzum, D. (2018) · IEEE Electron Device Letters · doi:10.1109/led.2018.2872434 · Cited by 49
“To overcome the limitation of the conventional architecture, crossbar array of resistive synaptic devices [1], [2] have been developed as a promising alternative to implement NNs in an energy-efficient manner.”
MethodKwon, D., Lim, S., Bae, J., Lee, S., Kim, H., Seo, Y.-T., Oh, S., Kim, J., Yeom, K., Park, B., et al. (2020) · Frontiers in Neuroscience · doi:10.3389/fnins.2020.00423 · Cited by 55
“For this reason, the performance of SNNs that adopt conversion is sensitive to unexpected variations of hardware and cannot save the power consumption required for training a weight (Kim H. et al., 2018; Yu, 2018).”
BackgroundMemTorch: A Simulation Framework for Deep Memristive Cross-Bar Architectures
Lammie, C., & Azghadi, M. R. (2020) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas45731.2020.9180810 · Cited by 50
Chen, B., Wang, X., Jiao, F., Ning, L., Huang, J., Xie, J., Zhang, S., Li, X., & Rao, F. (2023) · Advanced Science · doi:10.1002/advs.202301043 · Cited by 47
Memristive Quantized Neural Networks: A Novel Approach to Accelerate Deep Learning On-Chip
Zhang, Y., Cui, M., Shen, L., & Zeng, Z. (2019) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2019.2912205 · Cited by 50
“Neuro-inspired computing with emerging nonvolatile memory (NVM) has been investigated [47].”
BackgroundKim, S., Lee, Y., Kim, H., & Choi, S. (2019) · Nanoscale · doi:10.1039/c9nr08979a · Cited by 47
Mahata, C., Algadi, H., Ismail, M., Kwon, D., & Kim, S. (2021) · Journal of Material Science and Technology · doi:10.1016/j.jmst.2021.03.079 · Cited by 46
Low-Power Artificial Neural Network Perceptron Based on Monolayer MoS2
Marega, G. M., Wang, Z., Paliy, M., Giusi, G., Strangio, S., Castiglione, F., Callegari, C., Tripathi, M., Rađenović, A., Iannaccone, G., et al. (2022) · ACS Nano · doi:10.1021/acsnano.1c07065 · Cited by 48
Spiking neural networks for inference and learning: a memristor-based design perspective
Fouda, M. E., Kurdahi, F. J., Eltawil, A. M., & Neftci, E. (2020) · Elsevier eBooks · doi:10.1016/b978-0-08-102782-0.00019-8 · Cited by 53
Yoon, Y., Kim, Y., Hwang, W. S., & Shin, M. (2023) · Advanced Electronic Materials · doi:10.1002/aelm.202300098 · Cited by 52
A Novel Scalable Energy-Efficient Synaptic Device: Crossbar Ferroelectric Semiconductor Junction
Si, M., Luo, Y.-C., Chung, W., Bae, H., Zheng, D., Li, J., Qin, J., Qiu, G., Yu, S., & Ye, P. D. (2019) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm19573.2019.8993622 · Cited by 45
Self-citationBuilding Functional Memories and Logic Circuits with 2D Boron Nitride
Liu, X., Chen, J., Wang, Y., Han, S., & Zhou, Y. (2020) · Advanced Functional Materials · doi:10.1002/adfm.202004733 · Cited by 47
Optimized programming algorithms for multilevel RRAM in hardware neural networks
Milo, V., Anzalone, F., Zambelli, C., Pérez, E., Mahadevaiah, M. K., Ossorio, Ó. G., Olivo, P., Wenger, C., & Ielmini, D. (2021) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps46558.2021.9405119 · Cited by 50
“While these major advantages have been explored in neural network arrays in recent years [10]–[13], both device-to-device (D2D) and cycle-to-cycle (C2C) variations of resistance raise strong concerns in achieving high accuracy for weight programming in hardware arrays [14].”
BackgroundHsu, C.-L., Saleem, A., Singh, A., Kumar, D., & Tseng, T. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3112109 · Cited by 46
Kim, J. P., Kim, S. K., Park, S., Kuk, S., Kim, T.-Y., Kim, B. H., Ahn, S., Cho, Y., Jeong, Y., Choi, S., et al. (2023) · Nano Letters · doi:10.1021/acs.nanolett.2c03453 · Cited by 45
Memristive Devices for Neuromorphic Applications: Comparative Analysis
Erokhin, V. V. (2020) · BioNanoScience · doi:10.1007/s12668-020-00795-1 · Cited by 46
Ismail, M., Mahata, C., Abbas, H., Choi, C., & Kim, S. (2021) · Journal of Alloys and Compounds · doi:10.1016/j.jallcom.2020.158416 · Cited by 44
Recent progress on two-dimensional neuromorphic devices and artificial neural network
Tian, C., Wei, L., Li, Y., & Jiang, J. (2021) · Current Applied Physics · doi:10.1016/j.cap.2021.08.014 · Cited by 44
Enabling Training of Neural Networks on Noisy Hardware
Gokmen, T. (2021) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2021.699148 · Cited by 51
“Alternative to digital CMOS, hardware architectures composed of novel resistive cross-point device arrays have been proposed that can deliver signi fi cant power and speed bene fi ts for DNN training (Gokmen and Vlasov, 2016; Haensch et al., 2019; Burr et al., 2017; Burr et al., 2015; Yu, 2018).”
BackgroundA perspective on the physical scaling down of hafnia-based ferroelectrics
Park, J. Y., Lee, D. H., Park, G. H., Lee, J., Lee, Y. H., & Park, M. H. (2023) · Nanotechnology · doi:10.1088/1361-6528/acb945 · Cited by 43
Memcapacitor Crossbar Array with Charge Trap NAND Flash Structure for Neuromorphic Computing
Hwang, S., Yu, J., Song, M. S., Hwang, H., & Kim, H. (2023) · Advanced Science · doi:10.1002/advs.202303817 · Cited by 45
Dual-Functional Long-Term Plasticity Emulated in IGZO-Based Photoelectric Neuromorphic Transistors
He, Y., Nie, S., Liu, R., Jiang, S., Shi, Y., & Wan, Q. C. (2019) · IEEE Electron Device Letters · doi:10.1109/led.2019.2908727 · Cited by 45
Ostwal, V., Zand, R., DeMara, R. F., & Appenzeller, J. (2019) · IEEE Journal on Exploratory Solid-State Computational Devices and Circuits · doi:10.1109/jxcdc.2019.2956468 · Cited by 46
“memories (eNVMs) have attracted attention in the research community for their potential as synaptic elements [1].”
BackgroundFELIX: A Ferroelectric FET Based Low Power Mixed-Signal In-Memory Architecture for DNN Acceleration
Soliman, T., Laleni, N., Kirchner, T., Müller, F., Shrivastava, A., Kämpfe, T., Guntoro, A., & Wehn, N. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3529760 · Cited by 47
“Additionally, in such systems, the usage of of-chip memory, results in both substantial energy and latency [47].”
BackgroundHigh-Yield and Uniform NbO x -Based Threshold Switching Devices for Neuron Applications
Chen, P., Zhang, X., Wu, Z., Wang, Y., Zhu, J., Hao, Y., Feng, G., Sun, Y., Shi, T., Wang, M., et al. (2022) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2022.3161614 · Cited by 43
Recent Advances in Black Phosphorus‐Based Electronic Devices
Tan, W. C., Wang, L., Feng, X., Chen, L., Huang, L., Huang, X., & Ang, K. (2018) · Advanced Electronic Materials · doi:10.1002/aelm.201800666 · Cited by 42
Adaptive Weight Quantization Method for Nonlinear Synaptic Devices
Kwon, D., Lim, S., Bae, J., Lee, S., Kim, H., Kim, C.-H., Park, B., & Lee, J. (2018) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2018.2879821 · Cited by 43
“On-chip training schemes in HNNs have advantages of immunity to the device variation and reducing power consumption in weight training [2], [8], [9].”
BackgroundMethodCryogenic behavior of NbO2 based threshold switching devices as oscillation neurons
Wang, P., Khan, A. I., & Yu, S. (2020) · Applied Physics Letters · doi:10.1063/5.0006467 · Cited by 43
Self-citationImpact of Read Disturb on Multilevel RRAM based Inference Engine: Experiments and Model Prediction
Shim, W., Luo, Y., Seo, J.-S., & Yu, S. (2020) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps45951.2020.9129252 · Cited by 49
“Several emerging memories such as RRAM [1]-[6] and PCRAM [7]-[8], as well as mainstream memories such as SRAM [9] and FLASH [10]- [12] have been investigated for in-memory computing applications.”
BackgroundSelf-citationFerroelectric devices and circuits for neuro-inspired computing
Wang, P., & Yu, S. (2020) · MRS Communications · doi:10.1557/mrc.2020.71 · Cited by 45
Self-citationWu, J., Mo, F., Saraya, T., Hiramoto, T., & Kobayashi, M. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.3033831 · Cited by 44
“(b) Energy per forward propagation in inference, estimated through the modified neurosim framework [7].”
BackgroundMethodRecent Progress of Protein‐Based Data Storage and Neuromorphic Devices
Wang, J., Qian, F., Huang, S., Lv, Z., Wang, Y., Xing, X., Chen, M., Han, S., & Zhou, Y. (2020) · Advanced Intelligent Systems · doi:10.1002/aisy.202000180 · Cited by 43
IR-QNN Framework: An IR Drop-Aware Offline Training of Quantized Crossbar Arrays
Fouda, M. E., Lee, S., Lee, J., Kim, G. H., Kurdahi, F. J., & Eltawil, A. M. (2020) · IEEE Access · doi:10.1109/access.2020.3044652 · Cited by 44
Noh, J., Bae, H., Li, J., Luo, Y., Qu, Y., Park, T. J., Si, M., Chen, X., Charnas, A. R., Chung, W., et al. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3064783 · Cited by 44
“The on-chip learning was trained using the adaptive momentum estimation (ADAM) algorithm to perform handwritten digit classification for the MNIST data set [23]–[25].”
MethodSelf-citationModeling and Design of FTJs as Multi-Level Low Energy Memristors for Neuromorphic Computing
Fontanini, R., Segatto, M., Massarotto, M., Specogna, R., Driussi, F., Loghi, M., & Esseni, D. (2021) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2021.3120200 · Cited by 42
Baroni, A., Glukhov, A., Pérez, E., Wenger, C., Ielmini, D., Olivo, P., & Zambelli, C. (2022) · IEEE Transactions on Device and Materials Reliability · doi:10.1109/tdmr.2022.3182133 · Cited by 42
Mask Technique for Fast and Efficient Training of Binary Resistive Crossbar Arrays
Fouda, M. E., Lee, S., Lee, J., Eltawil, A., & Kurdahi, F. (2019) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2019.2927493 · Cited by 41
“While the previous work in this area typically uses a simplified model of crossbar arrays, in order to tackle large applications such as deep neural networks (DNNs), it is important to consider physical realities or RRAM’s non-idealities such as nonlinear and asymmetric potentiation and depression, device variability,…”
Background
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.