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 401–500 of 1,001
Katkar, P. K., Padalkar, N. S., Kumbhar, D. D., Patil, A. H., Sutar, S. S., Kadam, S. J., Kamat, R. K., Chun, S., & Dongale, T. D. (2022) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.2c00085 · Cited by 26
Han, J., Lee, J. W., Kim, Y., Kim, Y. B., Yun, S., Lee, S., Yu, J., Lee, K. J., Myung, H., & Choi, Y. (2023) · Advanced Science · doi:10.1002/advs.202302380 · Cited by 28
Neuromorphic Circuit of Classical and Operant Conditioning Based on Tunable Neural Circuitry Motifs
Guo, M. L., Kong, L., Dou, G., & Iu, H. H. (2024) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2024.3431588 · Cited by 28
“I NSPIRED by cognitive functions of biological brain, neuromorphic circuits have been extensively studied in order to achieve artificial intelligence by mimicking the mechanisms of neurons and synapses in the biological brain [1], [2].”
BackgroundAll‐Ferroelectric Spiking Neural Networks via Morphotropic Phase Boundary Neurons
Kim, J., Park, E. C., Shin, W., Koo, R., Im, J., Han, C., Lee, J., & Kwon, D. (2024) · Advanced Science · doi:10.1002/advs.202407870 · Cited by 25
Nath, S. K., Nandi, S. K., Helou, A. E., Liu, X., Li, S., Ratcliff, T., Raad, P. E., & Elliman, R. G. (2020) · Physical Review Applied · doi:10.1103/physrevapplied.13.064024 · Cited by 25
Halide perovskite based synaptic devices for neuromorphic systems
Beom, K., Fan, Z., Li, D., & Newman, N. (2022) · Materials Today Physics · doi:10.1016/j.mtphys.2022.100667 · Cited by 29
Liu, D., Zhou, H., Mao, W., Liu, J., Han, Y., Man, C., Wu, Q., Guo, Z., Huang, M., Luo, S., et al. (2022) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2022.3212314 · Cited by 26
“ReRAM-based CIM works are area-efficient and nonvolatile, which makes ReRAM array in CIM architecture extremely suitable for edge-computing applications [11], [12], [13], [14], [15], [16].”
BackgroundMemristor-Based Light-Weight Transformer Circuit Implementation for Speech Recognizing
Xiao, H., Zhou, Y., Gao, T., Duan, S., Chen, G., & Hu, X. (2023) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2023.3237582 · Cited by 26
“A proportional mapping scheme in which the conductance is proportional to the synaptic weights is primarily used [42], as it minimizes the complexity of the circuit required for mapping.”
MethodWang, Z., Zhou, X., Liu, X., Qiu, A., Gao, C., Yuan, Y., Jing, Y., Zhang, D., Li, W., Luo, H., et al. (2023) · Chip · doi:10.1016/j.chip.2023.100044 · Cited by 26
Kim, D.-H., Lee, J., Kim, Y., Yoo, H., & Lee, E. K. (2024) · Organic Electronics · doi:10.1016/j.orgel.2024.107076 · Cited by 25
Ultra-wide temperature electronic synapses based on self-rectifying ferroelectric memristors
Yang, N., Ren, Z., Hu, C.-Z., Guan, Z., Tian, B., Zhong, N., Xiang, P., Duan, C.-G., & Chu, J. (2019) · Nanotechnology · doi:10.1088/1361-6528/ab3c3d · Cited by 25
Park, Y., Kim, M., & Lee, J. (2020) · Carbon · doi:10.1016/j.carbon.2020.04.096 · Cited by 25
Novel Method Enabling Forward and Backward Propagations in NAND Flash Memory for On-Chip Learning
Lee, S., Yeom, G., Yoo, H.-N., Kim, H., Lim, S., Bae, J., Park, B., & Lee, J. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3081610 · Cited by 24
Quesada, E. P.-B., Mahadevaiah, M. K., Rizzi, T., Wen, J., Ulbricht, M., Krstic, M. D., Wenger, C., & Pérez, E. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3244509 · Cited by 24
“…(RRAMs), phase change memories (PCMs), magnetoresistive random access memories (MRAMs) and ferroelectric random access memories (FeRAMs) [7], [8], [9], [10] can boost the performance and efficiency of MAC operations computed directly within the memory units, utilizing basic circuit laws like…”
BackgroundLiu, X., Katti, K., & Jariwala, D. (2023) · Matter · doi:10.1016/j.matt.2023.03.016 · Cited by 24
Polarization and domains in wurtzite ferroelectrics: Fundamentals and applications
Fichtner, S., Schönweger, G., Lee, C., Yazawa, K., Gorai, P., & Brennecka, G. L. (2025) · Applied Physics Reviews · doi:10.1063/5.0249265 · Cited by 28
Du, 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
“both industry and academic in recent years [4], [5].”
BackgroundWang, P., Wang, Z., Sun, X., Hur, J. H., Datta, S., Khan, A. I., & Yu, S. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.3009956 · Cited by 25
“, activations from the previous layer) will activate multiple rows of the memory array as voltage, and the analog current along the columns will represent the weighted sum as the output vector [1].”
BackgroundSelf-citationKhan, S. A., & Kim, S. (2020) · RSC Advances · doi:10.1039/d0ra06389d · Cited by 24
Overview of the Role of Alloying Modifiers on the Performance of Phase Change Memory Materials
Kang, L., & Chen, L. (2020) · Journal of Electronic Materials · doi:10.1007/s11664-020-08590-0 · Cited by 24
XOR-CIM: compute-in-memory SRAM architecture with embedded XOR encryption
Huang, S., Jiang, H., Peng, X., Li, W., & Yu, S. (2020) · doi:10.1145/3400302.3415678 · Cited by 42
“Instead of transferring data between computing and storage units, the CIM architecture addresses the data movement problem by emerging the computation within memory arrays [3].”
Self-citationBenchmark of Ferroelectric Transistor-Based Hybrid Precision Synapse for Neural Network Accelerator
Luo, Y., Wang, P., Peng, X., Sun, X., & Yu, S. (2019) · IEEE Journal on Exploratory Solid-State Computational Devices and Circuits · doi:10.1109/jxcdc.2019.2925061 · Cited by 25
“Nowadays, intensive research and development efforts have been focused on deep learning accelerators with CMOS and beyond-CMOS technologies [1]–[6].”
BackgroundSelf-citationChoe, G., Shim, W., Wang, P., Hur, J. H., Khan, A. I., & Yu, S. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3068086 · Cited by 23
“In this context, the in-memory computing has received the spotlight as a candidate to overcome this bottleneck in neural networks by accelerating the computation [3].”
BackgroundSelf-citationFontanini, R., Segatto, M., Nair, K. S., Holzer, M., Driussi, F., Häusler, I., Koch, C. T., Dubourdieu, C., Deshpande, V. V., & Esseni, D. (2022) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2022.3175684 · Cited by 23
“The demand for energy efficient hardware platforms for artificial intelligence has raised a growing interest for memristive devices capable of implementing non volatile, multi–level conductance [1].”
BackgroundUltra-large dynamic range synaptic indium gallium zinc oxide transistors
Kwak, T., Kim, M.-J., Lee, W., Kim, E. C., Jang, J., Wang, G., Kim, T., Kim, Y. S., & Park, S. (2022) · Applied Materials Today · doi:10.1016/j.apmt.2022.101648 · Cited by 25
Robinson, D. A., Foster, M. E., Bennett, C. H., Bhandarkar, A., Webster, E. R., Celebi, A., Celebi, N., Fuller, E. J., Stavila, V., Spataru, C. D., et al. (2022) · Advanced Materials · doi:10.1002/adma.202207595 · Cited by 24
Kim, D., Lee, C. B., Park, K. K., Bang, H., Truong, P. L., Lee, J., Jeong, B. H., Kim, H., Won, S. M., Kim, D. H., et al. (2023) · ACS Nano · doi:10.1021/acsnano.3c05493 · Cited by 23
Mandal, R., Mandal, A., & Som, T. (2024) · Applied Materials Today · doi:10.1016/j.apmt.2024.102103 · Cited by 23
Kim, H., Bae, J., Lim, S., Lee, S., Seo, Y.-T., Kwon, D., Park, B., & Lee, J. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.09.099 · Cited by 26
Solution-Processed Multiterminal Artificial Synapses Based on Ion-Doped Solid Electrolytes
Park, M., Park, Y., & Lee, J. (2020) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.9b00788 · Cited by 25
SSM: a high-performance scheme for in situ training of imprecise memristor neural networks
Wang, Y., Wu, S., Tian, L., & Shi, L. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.04.130 · Cited by 28
Lee, S., Kwon, D., Kim, H., Yoo, H.-N., & Lee, J. (2020) · IEEE Access · doi:10.1109/access.2020.3004045 · Cited by 23
“Instead of a high-precision floating-point weight and activation, they enable a low-bit weight in synaptic devices and low-bit activation in neuron circuits, providing a promising solution to the implementation of neuromorphic systems [8], [12].”
BackgroundManipulation of the electrical behaviors of Cu/MXene/SiO2/W memristor
Wang, Y., Liu, X., Chen, Y., Xu, W., Liang, D., Gao, F., Zhang, M., Samanta, S., Gong, X., Lian, X., et al. (2019) · Applied Physics Express · doi:10.7567/1882-0786/ab4233 · Cited by 26
III-nitride memristors: materials, devices, and applications
Yang, Y., Li, H., & Hua, Q. (2025) · Materials Futures · doi:10.1088/2752-5724/ade5be · Cited by 22
An Ultralow Power Li x TiO 2 ‐Based Synaptic Transistor for Scalable Neuromorphic Computing
Nguyen, N., Schneegans, O., Salot, R., Lamy, Y., Giapintzakis, J., Mai, V. H., & Oukassi, S. (2022) · Advanced Electronic Materials · doi:10.1002/aelm.202200607 · Cited by 22
Memristive Circuit Implementation of Caenorhabditis Elegans Mechanism for Neuromorphic Computing
Chen, H., Hong, Q., Wang, Z., Wang, C., Zeng, X., & Zhang, J. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3250655 · Cited by 22
“development of digital electronics, is dying [1], [2], [3].”
BackgroundShen, G., Zhuge, C., Jiang, J., Fu, Y., Zheng, Y., Qin, Z., Wang, Q., & He, D. (2023) · Advanced Functional Materials · doi:10.1002/adfm.202309054 · Cited by 22
Digital image processing realized by memristor-based technologies
Wang, L., Meng, Q., Wang, H., Jiang, J., Wan, X., Liu, X., Lian, X., & Cai, Z. (2023) · Discover Nano · doi:10.1186/s11671-023-03901-w · Cited by 23
“Moreover, memristors have optimal write energy and standby power, where the majority of pulse-code modulated (PCM) devices and resistive random access memories (RRAM) have write energies of about 10–100 pJ and 100 fJ–10 pJ [47], respectively.”
BackgroundNonsilicon, Non-von Neumann Computing—Part I [Scanning the Issue]
Basu, S., Bryant, R. E., Micheli, G. D., Theis, T., & Whitman, L. J. (2018) · Proceedings of the IEEE · doi:10.1109/jproc.2018.2884780 · Cited by 22
Long Short-Term Memory Implementation Exploiting Passive RRAM Crossbar Array
Nikam, H., Satyam, S., & Sahay, S. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3133197 · Cited by 24
Two-dimensional materials enabled next-generation low-energy compute and connectivity
Pal, A., Agashiwala, K., Jiang, J., Zhang, D., Chavan, T., Kumar, A., Yeh, C., Cao, W., & Banerjee, K. (2021) · MRS Bulletin · doi:10.1557/s43577-022-00270-0 · Cited by 23
NV-BNN: An Accurate Deep Convolutional Neural Network Based on Binary STT-MRAM for Adaptive AI Edge
Chang, C.-C., Wu, M.-H., Lin, J., Li, C., Parmar, V., Lee, H.-Y., Wei, J., Sheu, S.-S., Suri, M., Chang, T., et al. (2019) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3316781.3317872 · Cited by 29
“This concept is being actively researched for constructing highly efficient hardware-defined neural networks (HNNs) on chips for accelerating DNNs [1, 2].”
BackgroundInfluentialArchitectural Design of 3D NAND Flash based Compute-in-Memory for Inference Engine
Shim, W., Jiang, H., Peng, X., & Yu, S. (2020) · International Symposium on Memory Systems (MEMSYS) · doi:10.1145/3422575.3422779 · Cited by 23
“Although all the duplication methods are applied on 3D NAND array, the total chip size is only 17.91 mm 2 which is significantly smaller than the CIM chips with other technologies (73.53 mm 2 for RRAM and over 100 mm 2 for SRAM).”
InfluentialSelf-citationA Memristor‐Based Bioinspired Multimodal Sensory Memory System for Sensory Adaptation of Robots
Pan, X., Wang, J., Deng, Z., Shuai, Y., Luo, W., Luo, W., Xie, Q., Xiao, Y., Tang, S. Y., Jiang, S., et al. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200031 · Cited by 21
Convolutional Echo‐State Network with Random Memristors for Spatiotemporal Signal Classification
Wang, S., Chen, H., Zhang, W., Li, Y., Wang, D., Shi, S., Zhao, Y., Loong, K. C., Chen, X., Dong, Y., et al. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200027 · Cited by 22
Review of neuromorphic computing based on NAND flash memory
Lee, S., & Lee, J.-H. (2024) · Nanoscale Horizons · doi:10.1039/d3nh00532a · Cited by 22
Implementing Binarized Neural Networks with Magnetoresistive RAM without Error Correction
Hirtzlin, T., Penkovsky, B., Klein, J., Locatelli, N., Vincent, A. F., Bocquet, M., Portal, J., & Querlioz, D. (2019) · IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH) · doi:10.1109/nanoarch47378.2019.181300 · Cited by 24
“Nevertheless, the most attractive implementations propose using RRAMs or ST-MRAMs for implementing the synaptic weights, in architectures that tightly integrate logic and memory [8], [10], [11], [14], [22], [25]– [27].”
BackgroundMethodA Variation Robust Inference Engine Based on STT-MRAM with Parallel Read-Out
Luo, Y., Peng, X., Hatcher, R., Rakshit, T., Kittl, J. A., Rodder, M. S., Seo, J.-S., & Yu, S. (2020) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas45731.2020.9181022 · Cited by 21
“I NTRODUCTION In-memory computing [1] is proposed to accelerate the intensive vector-matrix multiplication in DNN algorithms, where the multiplication is conducted in analog column current domain and the digital outputs are obtained after analog to digital converter (ADC).”
BackgroundSelf-citationWoo, S. Y., Kwon, D., Choi, N., Kang, W.-M., Seo, Y.-T., Park, M., Bae, J., Park, B., & Lee, J. (2020) · IEEE Access · doi:10.1109/access.2020.3036088 · Cited by 21
“Recently, hardware-based neural networks (HNNs) have emerged for use in neuromorphic systems to compute complex data efficiently [1]–[3].”
Ahn, H., Kim, Y., Seo, S., Lee, J., Lee, S., Oh, S., Kim, B., Park, J., Kang, S.-M., Kim, Y., et al. (2025) · Advanced Materials · doi:10.1002/adma.202418582 · Cited by 22
Chee, M. Y., Dananjaya, P. A., Lim, G. J., Du, Y., & Lew, W. S. (2022) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.2c11016 · Cited by 21
A CFMB STT-MRAM-Based Computing-in-Memory Proposal With Cascade Computing Unit for Edge AI Devices
Zhou, Y., Zhou, Z., Wei, Y., Yang, Z., Lin, X., Dai, C., Hao, L., Peng, C., Cai, H., & Wu, X. (2023) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2023.3327461 · Cited by 21
“This results in the necessity frequent data movement between SRAM and off-chip memories [17], [18], as depicted in Fig.1(a).”
BackgroundKim, Y., Kim, J., Kim, S. S., Kwon, Y. J., Kim, G. S., Jeon, J. W., Kwon, D. E., Yoon, J. H., & Hwang, C. S. (2020) · Advanced Intelligent Systems · doi:10.1002/aisy.201900116 · Cited by 17
Yan, Y. Q., Li, J. C., Chen, Y. T., Wang, X. Y., Cai, G., Park, H. W., Kim, J. H., Zhao, J., & Hwang, C. S. (2021) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.1c09436 · Cited by 22
Ali, T., Kühnel, K., Olivo, R. R., Lehninger, D., Müller, F., Lederer, M., Rudolph, M., Oehler, S., Mertens, K., Hoffmann, R., et al. (2021) · Electronic Materials · doi:10.3390/electronicmat2030024 · Cited by 21
“Today, the technological solutions for the Von-Neumann bottleneck cover broad concepts such as embedded nonvolatile memories [3], logic in-memory computing [4], or an alternative brain-inspired computing [5,6].”
BackgroundHan, G., Lee, C., Lee, J.-E., Seo, J., Kim, M.-J., Song, Y., Seo, Y.-H., & Lee, D. (2021) · Scientific Reports · doi:10.1038/s41598-021-02176-4 · Cited by 20
High On/Off Ratio Spintronic Multi‐Level Memory Unit for Deep Neural Network
Zhang, K., Jia, X., Cao, K., Wang, J., Zhang, Y., Lin, K., Chen, L., Feng, X., Zhang, Z., Zhang, Z., et al. (2022) · Advanced Science · doi:10.1002/advs.202103357 · Cited by 20
Lee, D. H., Kim, J. E., Cho, Y. H., Kim, S., Park, G. H., Choi, H., Lee, S. Y., Kwon, T., Kim, D. H., Jeong, M., et al. (2024) · Materials Horizons · doi:10.1039/d4mh00519h · Cited by 19
Towards scalable memristive hardware for spiking neural networks
Chen, P., Zhang, B., He, E., Xiao, Y., Liu, F., Lin, P., Wang, Z., & Pan, G. (2025) · Materials Horizons · doi:10.1039/d4mh01676a · Cited by 19
Design and Optimization of Non-Volatile Capacitive Crossbar Array for In-Memory Computing
Luo, Y.-C., Lu, A., Hur, J. H., Li, S., & Yu, S. (2021) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2021.3108148 · Cited by 19
“T O ACCELERATE the intensive vector-matrix multiplication (VMM) in deep learning algorithms, in-memory computing (IMC) [1] is proposed with parallel operations of multiplication and sum.”
Self-citationHigh-performance artificial neurons based on Ag/MXene/GST/Pt threshold switching memristors
Lian, X., Fu, J., Gao, Z., Gu, S., & Wang, L. (2022) · Chinese Physics B · doi:10.1088/1674-1056/ac673f · Cited by 19
Shu, H., Long, H., Sun, H., Li, B., Zhang, H., & Wang, X. (2022) · ACS Omega · doi:10.1021/acsomega.1c06864 · Cited by 19
Diffusion-assisted displacive transformation in Yttrium-doped Sb2Te3 phase change materials
Li, K., Liu, B., Zhou, J., Elliott, S. R., & Sun, Z. (2023) · Acta Materialia · doi:10.1016/j.actamat.2023.118809 · Cited by 19
The FAPbI3 perovskite memristor with a PMMA passivation layer as an artificial synapse
Wu, Y., Huang, H., Xu, C., Cao, X., Lei, Z., Zhang, J., Zhao, Y. L., Wei, A., & Liu, Z. (2023) · Applied Physics A · doi:10.1007/s00339-023-06632-y · Cited by 20
“Finally, the device was demonstrated to possess excellent long-term potentiation (LTP), long-term depression (LTD) and spike timing dependent plasticity (STDP).”
BackgroundInfluentialEmbedded Intelligence in the Internet-of-Things
Dick, R. P., Shang, L., Wolf, M., & Yang, S. (2019) · IEEE Design and Test · doi:10.1109/mdat.2019.2957352 · Cited by 20
“For additional information, see Yu’s survey article [108].”
InfluentialHybrid Analog-Digital Learning with Differential RRAM Synapses
Hirtzlin, T., Bocquet, M., Ernoult, M., Klein, J., Nowak, E., Vianello, E., Portal, J., & Querlioz, D. (2019) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm19573.2019.8993555 · Cited by 19
Verhelst, M., & Murmann, B. (2020) · The frontiers collection · doi:10.1007/978-3-030-18338-7_18 · Cited by 23
Localized Probing of Dielectric Breakdown in Multilayer Hexagonal Boron Nitride
Ranjan, A., O’Shea, S. J., Bosman, M., Raghavan, N., & Pey, K. L. (2020) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.0c17107 · Cited by 19
Dual-Functional Hybrid Selectorless RRAM and Selection Device for Memory Array Application
Chen, Y., Lee, J. K., & Lin, C. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3095438 · Cited by 19
Electrical model of multi-level bipolar Ta2O5/TaOx Bi-layered ReRAM
Chee, H. L., Kumar, T. N., & Almurib, H. A. F. (2019) · Microelectronics Journal · doi:10.1016/j.mejo.2019.104616 · Cited by 19
Li, W., Read, J., Jiang, H., & Yu, S. (2022) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2022.3223031 · Cited by 18
“compute-in-memory (CIM) has been extensively explored as a promising solution to alleviate the von Neumann bottle-neck that conventional platforms face and speed up MAC operations in DNNs [2], [3].”
BackgroundSelf-citationSo, J., Kim, T., Shin, J.-H., Kim, D., & Kim, F. S. (2023) · Macromolecular Research · doi:10.1007/s13233-023-00188-9 · Cited by 18
Ferroelectric Field-Effect Transistor Synaptic Device With Hafnium-Silicate Interlayer
Kim, S., Shin, W., Kim, M., Kwon, K. R., Yim, J., Kim, J., Han, C., Jeong, S., Park, E. C., You, J. W., et al. (2023) · IEEE Electron Device Letters · doi:10.1109/led.2023.3324695 · Cited by 18
“multidomain polarization switching can be realized to implement analog synapses [1], [2], [3], [4], [5], [6], [7], [8], [9].”
BackgroundIndependent Component Analysis Using RRAMs
Fouda, M. E., Neftci, E., Eltawil, A. M., & Kurdahi, F. J. (2018) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2018.2880734 · Cited by 18
“behaviors are characterized by nonlinear and asymmetric update dynamics, which are considered as major limitations to their large-scale deployment in neural networks [11].”
BackgroundOvercoming Crossbar Nonidealities in Binary Neural Networks Through Learning
Fouda, M. E., Lee, J., Eltawil, A. M., & Kurdahi, F. J. (2018) · IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH) · doi:10.1145/3232195.3232226 · Cited by 18
Flexible Oxide-Based Schottky Neuromorphic TFTs With Configurable Spiking Dynamic Functions
He, Y., Zhu, Y., Chen, C., Liu, R., Jiang, S., Zhu, L. Q., Shi, Y., & Wan, Q. C. (2020) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2020.3022814 · Cited by 19
“The hardware implementation for neuromorphic devices and systems holds a great promise for achieving brain-like computation [7].”
BackgroundFerroelectric, Analog Resistive Switching in Back‐End‐of‐Line Compatible TiN/HfZrO4/TiOx Junctions
Bégon‐Lours, L., Halter, M., Popoff, Y., & Offrein, B. J. (2020) · physica status solidi (RRL) - Rapid Research Letters · doi:10.1002/pssr.202000524 · Cited by 18
Hassanzadeh, P. (2021) · Life Sciences · doi:10.1016/j.lfs.2021.119272 · Cited by 18
Chen, A., Zhang, Z., Ma, G., Liu, N., Lin, C., Chen, W., Chang, T., & Wang, H. (2021) · IEEE Electron Device Letters · doi:10.1109/led.2021.3068823 · Cited by 18
“limitations of von Neumann architecture, but such circuits are complex and have low parallelism [5]–[8].”
BackgroundHe, K., Chakraborty, I., Wang, C., & Roy, K. (2022) · International Conference on Computer Aided Design · doi:10.1145/3508352.3549453 · Cited by 23
“To efficiently program the crossbar arrays, we adopt row-byrow closed-looped programming scheme [38].”
MethodSynaptic Transistor Arrays Based on PVA/Lignin Composite Electrolyte Films
Zhang, W., Li, J., Cheng, L., Shi, W., Lei, Y., Wen, S., Wang, F., Jiang, J., Wen, P., & Zhang, J. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3265940 · Cited by 18
“the efficiency of traditional computers [1].”
Skyrmion-Induced Memristive Magnetic Tunnel Junction for Ternary Neural Network
Pan, B., Zhang, D., Zhang, X., Wang, H., Bai, J., Yang, J., Zhang, Y., Kang, W., & Zhao, W. S. (2019) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2019.2913637 · Cited by 19
Self-Testing Analog Spiking Neuron Circuit
El-Sayed, S. A., Camuñas-Mesa, L. A., Linares-Barranco, B., & Stratigopoulos, H. (2019) · International Conference on Synthesis, Modeling, Analysis and Simulation Methods and Applications to Circuit Design · doi:10.1109/smacd.2019.8795234 · Cited by 18
“On the other hand, there are emerging architectures that use nonvolatile memory devices for implementing efficiently the synapses [9].”
BackgroundShim, W., & Yu, S. (2021) · IEEE Journal on Exploratory Solid-State Computational Devices and Circuits · doi:10.1109/jxcdc.2021.3057856 · Cited by 20
“Compute-in-memory (CIM) has been emerged as an alternative paradigm to overcome these challenges due to its high throughput and energy efficiency [2].”
BackgroundSelf-citationIn-Sensor-Memory Computing for Post-Von Neumann Intelligence: A Perspective
Tang, H., Yu, N., Min, P., Guo, R., & Zhang, G. (2026) · Nano-Micro Letters · doi:10.1007/s40820-026-02191-y · Cited by 21
“Since ISMC performance cannot be captured by isolated metrics like throughput alone, new standards need to encompass device physics, circuit behavior, and task-level accuracy to facilitate fair comparisons between disparate material systems and digital baselines [152].”
Lee, K.-H., Kwon, D., Woo, S. Y., Ko, J., Choi, W. Y., Park, B., & Lee, J. (2022) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2022.3207707 · Cited by 17
“power increase and long latency due to the bottleneck caused by data transfer between the processing unit and the memory unit [3], [4], [5].”
BackgroundElectrolyte Gated Synaptic Transistor based on an Ultra‐Thin Film of La 0.7 Sr 0.3 MnO 3
López, A., Tornos, J., Peralta, A., Barbero, I. E., Fernandez‐Canizares, F., Sánchez‐Santolino, G., Varela, M., Rivera, A., Camarero, J., Leon, C., et al. (2023) · Advanced Electronic Materials · doi:10.1002/aelm.202300007 · Cited by 17
On-Chip Learning in Vertical NAND Flash Memory Using Forward–Forward Algorithm
Park, S., Ko, J., Lee, I.-S., Koo, R., Kim, J., Yang, Y., Kwon, D., Kim, J., & Lee, J. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3392170 · Cited by 17
“However, the conventional von Neumann architecture, which separates memory and processor, suffers from significant time and energy consumption in data movement between the two components [4], [5].”
BackgroundZhao, B., Xin, Z., Wang, Y., Wu, C., Wang, W., Shi, R., Peng, R., Wu, Y., Xu, L., Pan, T., et al. (2025) · Matter · doi:10.1016/j.matt.2025.102044 · Cited by 18
Influence of Compact Memristors’ Stability on Machine Learning
Zhu, R., Ye, S., Tang, Z., Lin, P., Huang, Q., Wang, H., He, J. L., & Chang, S. (2019) · IEEE Access · doi:10.1109/access.2019.2909295 · Cited by 17
“Basing on this model, to unveil the effect of memristors’ character on neural networks, we point out a detail description of the most common instabilities in memristors, device to device variations and cycle to cycle variations, [22].”
MethodStorage Reliability of Multi-bit Flash Oriented to Deep Neural Network
Xiang, Y., Kang, J., Huang, P. C., Yang, H., Wang, K. L., Han, R., Shen, W. S., Feng, Y., LIU, C. L. Q., & Liu, X. Y. (2019) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm19573.2019.8993508 · Cited by 17
“To improve the in-memory computing capacity and achieve cost-per-bit reduction, multibit data storage is crucial [4].”
BackgroundAn Efficient 3D ReRAM Convolution Processor Design for Binarized Weight Networks
Kim, B., Hanson, E., & Li, H. H. (2021) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2021.3067840 · Cited by 17
“By using ReRAM as a digital device, it also removes the ambiguity coming from the device variations, a representative non-ideal property of ReRAM [6].”
MethodKim, M., Hwang, S., Bang, S., Kim, T., Lee, D. K., Ansari, M. H. R., Cho, S., & Park, B. (2021) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2021.3099769 · Cited by 17
Hopfield neural network in magnetic textures with intrinsic Hebbian learning
Yu, W., Xiao, J., & Bauer, G. E. W. (2021) · Physical review. B./Physical review. B · doi:10.1103/physrevb.104.l180405 · Cited by 17
Exploiting Non-idealities of Resistive Switching Memories for Efficient Machine Learning
Yon, V., Amirsoleimani, A., Alibart, F., Melko, R. G., Drouin, D., & Beilliard, Y. (2022) · Frontiers in Electronics · doi:10.3389/felec.2022.825077 · Cited by 18
“The dynamics of the resistance programming can also pose a challenge for NN training (Sidler et al., 2016; Woo and Yu, 2018).”
BackgroundPost-silicon nano-electronic device and its application in brain-inspired chips
Lv, Y., Chen, H., Wang, Q., Li, X., Xie, C., & Song, Z. (2022) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2022.948386 · Cited by 18
“Some examples of state-of-the-art based on RRAM are given in the literature, all of which show bidirectional graded conductance tuning under the same programming voltage pulse (Mulaosmanovic et al., 2017; Yu, 2018).”
BackgroundWoo, J., & Yu, S. (2018) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2018.2875937 · Cited by 19
“Among the reported emerging devices, a resistive random access memory (RRAM) has been considered promising because of multiple conductance states corresponding to the weights [5], which can be also analogously adjusted by identical pulses from the wordline and bitline.”
BackgroundSelf-citationHighly Reliable Inference System of Neural Networks Using Gated Schottky Diodes
Lim, S., Kwon, D., Eum, J.-H., Lee, S., Bae, J., Kim, H., Kim, C.-H., Park, B., & Lee, J. (2019) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2019.2913146 · Cited by 18
“Various electronic devices such as resistive change memory (RRAM), phase change memory (PCRAM), and FET-based devices as synaptic devices have been studied to implement the HNN [2], [3].”
MethodAnalog Resistive Switching Phenomena in Titanium Oxide Thin-Film Memristive Devices
Islam, K., Sultana, R., & Mroczyński, R. (2025) · Materials · doi:10.3390/ma18153454 · Cited by 16
“This bidirectional modulation corresponds to LTP (increased conductance) and LTD (decreased conductance), both of which are crucial for mimicking the adaptive learning processes observed in computational neuroscience, neural networks, and biological systems [74].”
Read, J., Lee, M., Huang, W.-H., Luo, Y.-C., Lu, A., & Yu, S. (2026) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2026.3674448 · Cited by 21
“Analog CIM (ACIM) leverages memory cell physics for current, charge, or time-domain computation, where weights encode as conductance or capacitance, and column currents/charges are summed to yield MAC results [4], [5].”
Self-citation
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.