High-Throughput In-Memory Computing for Binary Deep Neural Networks With Monolithically Integrated RRAM and 90-nm CMOS
Open accessThe DOI matches its Crossref record; 47 of the 55 references it cites were found in a registry or an index, and 8 are not checked yet.
Abstract
Deep neural network (DNN) hardware designs have been bottlenecked by conventional memories, such as SRAM due to density, leakage, and parallel computing challenges. Resistive devices can address the density and volatility issues but have been limited by peripheral circuit integration. In this work, we present a resistive RAM (RRAM)-based in-memory computing (IMC) design, which is fabricated in 90-nm CMOS with monolithic integration of RRAM devices. We integrated a 128 × 64 RRAM array with CMOS peripheral circuits, including row/column decoders and flash analog-to-digital converters (ADCs), which collectively become a core component for scalable RRAM-based IMC for large DNNs. To maximize IMC parallelism, we assert all 128 wordlines of the RRAM array simultaneously, perform analog computing along the bitlines, and digitize the bitline voltages using ADCs. The resistance distribution of low-resistance states is tightened by an iterative write-verify scheme. Prototype chip measurements demonstrate high binary DNN accuracy of 98.5% for MNIST and 83.5% for CIFAR-10 data sets, with 24 TOPS/W and 158 GOPS. This represents 22.3× and 10.1× improvements in throughput and energy-delay product (EDP), respectively, compared with the state-of-the-art literature, which can enable intelligent functionalities for area-/energy-constrained edge computing devices.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“So far, several groups and companies have claimed the realization of hybrid CMOS/memristor implementations of ANNs53–61, —from now on, memristive ANNs— with performance that is superior to that of fully-CMOS counterparts.”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [61]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Some groups prefer to incorporate a transistor in series to each MIM cell to obtain a better control over the currents through the device (i.e., improve conductance controllability and minimize sneak path currents)53–55,60,61,102,113,167,194,218–222,).”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [61]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
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×
“DEEP neural networks (DNNs) have been very successful in large-scale recognition and classification tasks [1]–[5], while state-of-the-art deep learning algorithms tend to present very deep and large network models [1]–[3].”
BackgroundGradient-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 12×
“We scanned in 2,000 test input vectors from MNIST BNN benchmark, and obtained a 2-”
MethodInfluentialDensely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
Mastering 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×
“DEEP neural networks (DNNs) have been very successful in large-scale recognition and classification tasks [1]–[5], while state-of-the-art deep learning algorithms tend to present very deep and large network models [1]–[3].”
BackgroundGoogle's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Wu, Y., Schuster, M., Chen, Z., Le, Q. V., Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., et al. (2016) · arXiv · doi:10.48550/arxiv.1609.08144 · Cited by 5,319 · cited 1×
“This is especially true for embedded hardware applications, such as autonomous driving 6 , machine translation 7 , and smart wearable devices 8 , where severe constraints exist in performance, power, and area.”
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 7×
“Conventional binary RRAMs cannot effectively represent the positive and negative weight values (+1 and −1) in recent BNNs [8], [9] because the LRS and HRS values of binary RRAM devices are both positive.”
BackgroundMethodResultInfluentialA million spiking-neuron integrated circuit with a scalable communication network and interface
Merolla, P. A., Arthur, J. V., Alvarez-Icaza, R., Cassidy, A. S., Sawada, J., Akopyan, F., Jackson, B. L., Imam, N., Guo, C., Nakamura, Y., et al. (2014) · Science · doi:10.1126/science.1254642 · Cited by 4,147
Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks
Chen, Y.-H., Krishna, T., Emer, J. S., & Sze, V. (2017) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2016.2616357 · Cited by 3,247 · cited 13×
“An RRAM macro with multi-level sense amplifiers in 55nm CMOS logic process was recently reported 30 , but a relatively low accuracy of 81.83% accuracy for CIFAR-10 dataset was achieved with binary/ternary precision, and only 9 WLs are asserted simultaneously in the 256×512 sub-array, which limits further parallelism.”
BackgroundMethodInfluential
Cited by
Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey
Deng, B. L., Li, G., Han, S., Shi, L., & Xie, Y. (2020) · Proceedings of the IEEE · doi:10.1109/jproc.2020.2976475 · Cited by 996
“For the current to voltage (I2V) converters, they can be implemented either explicitly [258], [268] or implicitly [269] in different designs.”
Hardware implementation of memristor-based artificial neural networks
Aguirre, F., Sebastian, A., Le Gallo, M., Song, W., Wang, T., Yang, J. J., Lu, W., Chang, M.-F., Ielmini, D., Yang, Y., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-45670-9 · Cited by 564
“So far,several groups and companies have claimed the realization of hybrid CMOS/memristor implementations of ANNs 53 – 61 , — from now on, memristive ANNs — with performance that is superior to that of fully-CMOS counterparts.”
Memristor‐Based Neuromorphic Chips
Duan, X., Cao, Z., Gao, K., Yan, W., Sun, S., Zhou, G., Wu, Z., Ren, F., & Sun, B. (2024) · Advanced Materials · doi:10.1002/adma.202310704 · Cited by 432
Compute-in-Memory Chips for Deep Learning: Recent Trends and Prospects
Yu, S., Jiang, H., Huang, S., Peng, X., & Lu, A. (2021) · IEEE Circuits and Systems Magazine · doi:10.1109/mcas.2021.3092533 · Cited by 391
Self-citationRecent Advances and Future Prospects for Memristive Materials, Devices, and Systems
Song, M.-K., Kang, J.-H., Zhang, X., Ji, W., Ascoli, A., Messaris, I., Demirkol, A. S., Dong, B., Aggarwal, S., Wan, W., et al. (2023) · ACS Nano · doi:10.1021/acsnano.3c03505 · Cited by 300
Self-citationLe Gallo, M., Khaddam-Aljameh, R., Stanisavljevic, M., Vasilopoulos, A., Kersting, B., Dazzi, M., Karunaratne, G., Brändli, M., Singh, A., Müller, S. M., et al. (2023) · Nature Electronics · doi:10.1038/s41928-023-01010-1 · Cited by 310
Wurtzite and fluorite ferroelectric materials for electronic memory
Kim, K.-H., Karpov, I., Olsson, R. H., & Jariwala, D. (2023) · Nature Nanotechnology · doi:10.1038/s41565-023-01361-y · Cited by 283
Ultra-fast switching memristors based on two-dimensional materials
Teja Nibhanupudi, S. S., Roy, A., Veksler, D., Coupin, M., Matthews, K. C., Disiena, M., Ansh, Singh, J. V., Gearba-Dolocan, I. R., Warner, J., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-46372-y · Cited by 227
Mixed-Signal Computing for Deep Neural Network Inference
Murmann, B. (2021) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2020.3020286 · Cited by 160
“The natural solution is to process the sum in the current domain (via a transimpedance amplifier, integrator [46], or current-mode comparator [47]), but presumably more efficient voltage-domain readouts are also possible [48]–[50].”
BackgroundMethodRRAM for Compute-in-Memory: From Inference to Training
Yu, S., Shim, W., Peng, X., & Luo, Y. (2021) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3072200 · Cited by 171
“Recently, silicon macros of RRAM arrays integrated with peripheral circuits such as ADCs have also been demonstrated at 130nm [15], 90nm [16], 65nm [17], 55nm [18] and 22nm [19].”
BackgroundSelf-citationHänsch, W., Raghunathan, A., Roy, K., Chakrabarti, B., Phatak, C., Wang, C., & Guha, S. (2022) · Advanced Materials · doi:10.1002/adma.202204944 · Cited by 167
Exploring the Connection Between Binary and Spiking Neural Networks
Lu, S. K., & Sengupta, A. (2020) · Frontiers in Neuroscience · doi:10.3389/fnins.2020.00535 · Cited by 134
“Note that a dummy column will be required for referencing purposes of sense amplifiers interfaced with the array [29].”
BackgroundYin, S., Kim, Y., Han, X., Barnaby, H., Yu, S., Luo, Y., He, W., Sun, X., Kim, J.-J., & Seo, J.-S. (2019) · IEEE Micro · doi:10.1109/mm.2019.2943047 · Cited by 90
Self-citationMemory applications from 2D materials
Chiang, C., Ostwal, V., Wu, P., Pang, C., Zhang, F., Chen, Z., & Appenzeller, J. (2021) · Applied Physics Reviews · doi:10.1063/5.0038013 · Cited by 88
Li, W., Sun, X., Huang, S., Jiang, H., & Yu, S. (2022) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2022.3163197 · Cited by 77
Self-citationYoon, J., Chang, M., Khwa, W.-S., Chih, Y.-D., Chang, M., & Raychowdhury, A. (2022) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2022.3141370 · Cited by 71
“RBLs [23], [24].”
BackgroundByun, K., Choi, I., Kwon, S., Kim, Y., Kang, D., Cho, Y. W., Yoon, S. K., & Kim, S. (2022) · Advanced Materials Technologies · doi:10.1002/admt.202200884 · Cited by 68
Yoon, J., Chang, M., Khwa, W.-S., Chih, Y.-D., Chang, M., & Raychowdhury, A. (2021) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2021.3101209 · Cited by 68
“A diode-connected current source used in the current-sensing RD can provide the nonlinear BL current even in voltage-sensing RD [31].”
Toward memristive in-memory computing: principles and applications
Bao, H., Zhou, H., Li, J., Pei, H.-Z., Tian, J., Yang, L., Ren, S., Tong, S.-Q., Li, Y., He, Y., et al. (2022) · Frontiers of Optoelectronics · doi:10.1007/s12200-022-00025-4 · Cited by 79
Benchmarking In-Memory Computing Architectures
Shanbhag, N. R., & Roy, S. K. (2022) · IEEE Open Journal of the Solid-State Circuits Society · doi:10.1109/ojsscs.2022.3210152 · Cited by 57
“, 90% of the total bank area in [64], which further reduces their compute density (see Fig.”
BackgroundAutomatic diagnosis of COVID-19 with MCA-inspired TQWT-based classification of chest X-ray images
Jyoti, K., Sushma, S., Yadav, S. K., Kumar, P., Pachori, R. B., & Mukherjee, S. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.106331 · Cited by 51
“Various studies on memristor-based accelerator architectures [22] and memristor-based architectures for neuromorphic applications [23] have been previously published [24].”
BackgroundMethodIn-Memory Computing for Machine Learning and Deep Learning
Lepri, N., Glukhov, A., Cattaneo, L., Farronato, M., Mannocci, P., & Ielmini, D. (2023) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2023.3265875 · Cited by 59
“As for the fully digital approach, the XNOR gate can adopt the 1T1R cell [137], [140] with various differential techniques with NVM [139], [142], [148] or SRAM, whose output result is typically stored in a capacitor as a binary charge quantity [145], [149], [150].”
MethodAn Overview of Computing-in-Memory Circuits With DRAM and NVM
Kim, S., & Yoo, H. (2023) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2023.3333851 · Cited by 52
“Therefore, [26] uses flash ADC with non-linear reference voltage for sensing non-linear voltage transfer function with linear quantization.”
BackgroundMethodInfluentialLow-Power Memristor-Based Computing for Edge-AI Applications
Singh, A., Diware, S., Gebregiorgis, A., Bishnoi, R., Catthoor, F., Joshi, R., & Hamdioui, S. (2021) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas51556.2021.9401226 · Cited by 40
Brain-inspired computing systems: a systematic literature review
Zolfagharinejad, M., Alegre-Ibarra, U., Chen, T., Kinge, S., & van der Wiel, W. G. (2024) · The European Physical Journal B · doi:10.1140/epjb/s10051-024-00703-6 · Cited by 39
CMOS-integrated nanoscale memristive crossbars for CNN and optimization acceleration
Li, C., Ignowski, J., Sheng, X., Wessel, R., Jaffe, B., Ingemi, J., Graves, C., & Strachan, J. P. (2020) · 2020 IEEE International Memory Workshop (IMW) · doi:10.1109/imw48823.2020.9108112 · Cited by 37
Deaville, P., Zhang, B., Chen, L.-Y., & Verma, N. (2021) · ESSCIRC 2021 - IEEE 47th European Solid State Circuits Conference (ESSCIRC) · doi:10.1109/esscirc53450.2021.9567807 · Cited by 36
“While recent IMCs works have targeted ReRAM-based eNVM for row-parallel operation [1-4], IMC IP is required in the range of technologies suited for different application needs.”
BackgroundComprehending In-memory Computing Trends via Proper Benchmarking
Shanbhag, N. R., & Roy, S. K. (2022) · Custom Integrated Circuits Conference · doi:10.1109/cicc53496.2022.9772817 · Cited by 36
SRIF: Scalable and Reliable Integrate and Fire Circuit ADC for Memristor-Based CIM Architectures
Singh, A., Lebdeh, M. A., Gebregiorgis, A., Bishnoi, R., Joshi, R., & Hamdioui, S. (2021) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3061214 · Cited by 30
“Several other ADC designs such as [22], [26] involve flash ADC-like approach, which are fast but are significantly expensive in terms of area and power consumption.”
BackgroundMonolithically Integrated Complementary Ferroelectric FET XNOR Synapse for the Binary Neural Network
Hwang, J., Joh, H., Kim, C., Ahn, J., & Jeon, S. (2024) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.3c13945 · Cited by 29
Ling, Y., Wang, Z., Yu, Z., Bao, S., Yang, Y., Bao, L., Sun, Y., Cai, Y., & Huang, R. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3266186 · Cited by 29
“Our simulation is based on practical RRAM products and incorporates realistic configurations, such as the use of TiN electrodes as opposed to materials like Au and Pt which can cause impurities in the CMOS production line [30], [31], [32].”
MethodKwon, D., Woo, S. Y., Lee, K.-H., Hwang, J., Kim, H., Park, S., Shin, W., Bae, J., Kim, J., & Lee, J. (2023) · Science Advances · doi:10.1126/sciadv.adg9123 · Cited by 29
“High power consumption and a large area of analog-to-digital converters (ADCs) induce ADCs to be shared in NC blocks, lowering system efficiency ( 38 , 39 ).”
Krestinskaya, O., Zhang, L., & Salama, K. N. (2023) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2023.3293026 · Cited by 29
“[35] RRAM 90nm 1, 1, 1 9-layer CNN, 5-layer ANN Fabricated XNOR-RRAM prototype chip”
MethodAn Analysis of Components and Enhancement Strategies for Advancing Memristive Neural Networks
Park, H., Han, J., Yim, S., Shin, D. H., Park, T. W., Woo, K. S., Lee, S. H., Cho, J. M., Kim, H. W., Park, T., et al. (2025) · Advanced Materials · doi:10.1002/adma.202412549 · Cited by 27
A near-threshold memristive computing-in-memory engine for edge intelligence
Wang, L., Li, W., Zhou, Z., An, J., Ye, W., Li, Z., Gao, H., Hu, H., Liu, J., Chen, X., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-61025-4 · Cited by 38
Lefebvre, M., & Bol, D. (2022) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2022.3172647 · Cited by 26
“1× 2× 4× 8× M1_CALIB[0] [1] [2] [3] MSW Calibration circuitry Legend”
BackgroundNeuromorphic Computing Based on CMOS-Integrated Memristive Arrays: Current State and Perspectives
Mikhaylov, A. N., Gryaznov, E. G., Koryazhkina, M. N., Bordanov, I., Shchanikov, S., Telminov, O. A., & Kazantsev, V. (2023) · Supercomputing Frontiers and Innovations · doi:10.14529/jsfi230206 · Cited by 36
“Various circuits of adaptive compensation of large or small current values in polled memristors are used, for example, the voltage clamp control circuit [59].”
MethodYang, J., Mao, R., Jiang, M., Cheng, Y., Sun, P.-S. V., Dong, S., Pedretti, G., Sheng, X., Ignowski, J., Li, H., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-56254-6 · Cited by 38
Venu, B., Yellampalli, S. S., & Vaddi, R. (2023) · Microelectronics Journal · doi:10.1016/j.mejo.2023.105867 · Cited by 25
Field-Programmable Gate Array Architecture for Deep Learning: Survey and Future Directions
Boutros, A., Arora, A., & Betz, V. (2025) · Proceedings of the IEEE · doi:10.1109/jproc.2025.3623023 · Cited by 55
Progressive RESET induced by Joule heating in hBN RRAMs
Zhuang, P., Ma, W., Liu, J., Cai, W., & Lin, W. (2021) · Applied Physics Letters · doi:10.1063/5.0040902 · Cited by 24
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
“However, SRAM occupies a large area ( ∼ 150 F 2 per bitcell) and on-chip SRAM capacity ( ∼ a few megabytes) is insufficient to accommodate a large number of parameters [7].”
BackgroundHardware for Deep Learning Acceleration
Song, C., Ye, C., Sim, Y., & Jeong, D. S. (2024) · Advanced Intelligent Systems · doi:10.1002/aisy.202300762 · Cited by 29
Dual-configuration in-memory computing bitcells using SiOx RRAM for binary neural networks
Kingra, S. K., Parmar, V., Negi, S., Bricalli, A., Piccolboni, G., Regev, A., Nodin, J.-F., Molas, G., & Suri, M. (2022) · Applied Physics Letters · doi:10.1063/5.0073284 · Cited by 22
Deaville, P., Zhang, B., & Verma, N. (2024) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2024.3386462 · Cited by 21
“IMC based on embedded non-volatile memory (eNVM), such as resistive RAM (ReRAM) [6], MRAM [7], and phase-change memory (PCM) [8], [9], has been of high interest due to bit-cell density scalability and low-leakage retention for low-power edge applications.”
BackgroundMethodInfluentialGlobal-Gate Controlled One-Transistor One-Digital-Memristor Structure for Low-Bit Neural Network
Huang, M., Zhao, G., Wang, X., Zhang, W., Coquet, P., Tay, B. K., Zhong, G., & Li, J. (2020) · IEEE Electron Device Letters · doi:10.1109/led.2020.3037203 · Cited by 21
An In-Memory Analog Computing Co-Processor for Energy-Efficient CNN Inference on Mobile Devices
Elbtity, M., Singh, A. S., Reidy, B., Guo, X., & Zand, R. (2021) · IEEE Computer Society Annual Symposium on VLSI · doi:10.1109/isvlsi51109.2021.00043 · Cited by 21
“Some of the resistive memory technologies that have been used to realize IMC systems are resistive random-access memory (ReRAM) [4], [5], phase-change memory (PCM) [6], and magnetoresistive random-access memory (MRAM) [7].”
BackgroundSiO 2 Fin-Based Flash Synaptic Cells in AND Array Architecture for Binary Neural Networks
Lee, S., Kim, H., Lee, S., Park, B., & Lee, J. (2021) · IEEE Electron Device Letters · doi:10.1109/led.2021.3125966 · Cited by 22
Mayahinia, M., Singh, A., Bengel, C., Wiefels, S., Lebdeh, M. A., Menzel, S., Wouters, D. J., Gebregiorgis, A., Bishnoi, R., Joshi, R., et al. (2022) · ACM Journal on Emerging Technologies in Computing Systems · doi:10.1145/3451212 · Cited by 21
Kim, J., Kornijčuk, V., Ye, C., & Jeong, D. S. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.3027583 · Cited by 20
“To date, efforts to alleviate the complexity have been made using emerging devices, such as resistive random access memory arrays, which can theoretically reduce the complexity to O ( n ) when parallel addressing word lines and even O ( 1 ) when parallel addressing word and bit lines [5], [6].”
BackgroundCurrent opinions on memristor-accelerated machine learning hardware
Jiang, M., Xu, Y., Li, Z., & Li, C. (2025) · Current Opinion in Solid State and Materials Science · doi:10.1016/j.cossms.2025.101226 · Cited by 28
Jeong, S., Kim, J., Jeong, M., & Lee, Y. (2022) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2022.3165352 · Cited by 22
Xbar-Partitioning: A Practical Way for Parasitics and Noise Tolerance in Analog IMC Circuits
Amin, M. H., Elbtity, M., & Zand, R. (2022) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2022.3222966 · Cited by 21
“1109/JETCAS.2022.3222966 memory (RRAM) [4], phase-change memory (PCM) [5], magnetoresistive random-access memory (MRAM) [6], and conductive bridging random access memory (CBRAM) [7].”
BackgroundInfluentialNeuromorphic Accelerator for Spiking Neural Network Using SOT-MRAM Crossbar Array
Verma, G., Nisar, A., Dhull, S., & Kaushik, B. K. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3317357 · Cited by 23
“Each MLSA compares the RBL voltage of the selected column with a reference voltage [28], [29].”
BackgroundVasilopoulos, A., Büchel, J., Kersting, B., Lammie, C., Brew, K. W., Choi, S., Philip, T. M., Saulnier, N., Narayanan, V., Gallo, M. L., et al. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3321014 · Cited by 19
“Although Diff-N unit-cells increase the crossbar area, the impact on energy consumption can be limited if device power consumption is much lower than that of peripheral circuits, which is typically the case in current AIMC chips [15], [16], [17].”
BackgroundVerhelst, M., & Murmann, B. (2020) · The frontiers collection · doi:10.1007/978-3-030-18338-7_18 · Cited by 23
An SBT-memristor-based crossbar memory circuit*
Guo, M. L., Liu, R.-Y., Dou, M.-L., & Dou, G. (2021) · Chinese Physics B · doi:10.1088/1674-1056/abd7dc · Cited by 16
Unlocking wordline-level parallelism for fast inference on RRAM-based DNN accelerator
Park, Y., Lee, S.-Y., Shin, H., Heo, J., Ham, T. J., & Lee, J. W. (2020) · doi:10.1145/3400302.3415664 · Cited by 22
“Technically, RRAM array computation itself happens in a very short time (often less than five nanoseconds [32, 35]).”
BackgroundBitS-Net: Bit-Sparse Deep Neural Network for Energy-Efficient RRAM-Based Compute-In-Memory
Karimzadeh, F., Yoon, J., & Raychowdhury, A. (2022) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2022.3145687 · Cited by 21
“However, the parallel resistance drastically decreases over N.RRAM such that a sufficient sampling margin in readout cannot be secured even employing diode-connected current sources [41].”
BackgroundLi, 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
“Also, uniquely calibrated ADC references [20] are crucial to minimize the offset effects.”
BackgroundSelf-citationFalcone, D. F., Menzel, S., Stecconi, T., Galetta, M., Porta, A. L., Offrein, B. J., & Bragaglia, V. (2024) · Nanoscale Horizons · doi:10.1039/d4nh00072b · Cited by 17
A 65-nm RRAM Compute-in-Memory Macro for Genome Processing
Zhang, F., Sridharan, A., He, W., Yeo, I., Liehr, M., Zhang, W., Cady, N. C., Cao, Y. K., Seo, J.-S., & Fan, D. (2024) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2024.3396429 · Cited by 16
Self-citationNoh, K., Kwak, H., Son, J., Kim, S., Um, M., Kang, M., Kim, D., Ji, W., Lee, J.-Y., Jo, H., et al. (2024) · Science Advances · doi:10.1126/sciadv.adl3350 · Cited by 16
Kong, S., & Shim, W. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3447612 · Cited by 14
“…[9], [10], [11], [12], [13] have been investigated as candidates for PIM applications, such as static random access memory (SRAM) [2], [3], [4], resistive random access memory (RRAM) [5], [6], [7], phase change memory (PCM) [8], [9], and ferroelectric field-effect transistor (FeFET) [10], [11].”
BackgroundUltrafast and Energy-Efficient Ferrimagnetic XNOR Logic Gates for Binary Neural Networks
Wang, G., Zhang, Y., Zhang, Z., Zheng, Z., Zhang, K., Wang, J., Klein, J., Ravelosona, D., & Zhao, W. S. (2021) · IEEE Electron Device Letters · doi:10.1109/led.2021.3062382 · Cited by 15
“Based on the emerging non-volatile memories (NVMs), such as resistive random-access memory (RRAM) [6]–[8], magnetic RAM (MRAM), [9]–[11] and domain wall (DW) based race-track memory [12]–[14], BNNs can further take advantages of in-memory-computing (IMC) architecture by performing bit-operations…”
BackgroundInterconnect Parasitics and Partitioning in Fully-Analog In-Memory Computing Architectures
Amin, M. H., Elbtity, M., & Zand, R. (2022) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas48785.2022.9937884 · Cited by 20
“Accuracy comparisons show that a ∼55% accuracy drop for the non-ideal IMC architecture with HP = [7, 2, 2] and VP = [2, 2, 1] partitions can reduce to less than 1% accuracy drop for the highly-partitioned scenario with HP = [16, 8, 8] and VP = [8, 8, 1] partitions.”
The design of analogue in-memory computing tiles
Singh, A., Gallo, M. L., Vasilopoulos, A., Luquin, J., Narayanan, P., Burr, G. W., & Sebastian, A. (2025) · Nature Electronics · doi:10.1038/s41928-025-01537-5 · Cited by 21
“The flash ADC compares the analogue voltage with all possible reference levels in a single cycle using multiple comparators 45,60,61 .”
Mapping Binary ResNets on Computing-In-Memory Hardware with Low-bit ADCs
Kim, Y., Kim, H., Park, J., Oh, H., & Kim, J. (2021) · Design, Automation, and Test in Europe · doi:10.23919/date51398.2021.9474056 · Cited by 13
He, W., Meng, J., Gonugondla, S. K., Yu, S., Shanbhag, N. R., & Seo, J.-S. (2023) · Design, Automation, and Test in Europe · doi:10.23919/date56975.2023.10137266 · Cited by 12
“1 shows the high-level overview of the RRAM prototype chip reported in [18], [19].”
BackgroundMethodInfluentialSelf-citationMulti-bit MRAM based high performance neuromorphic accelerator for image classification
Verma, G., Soni, S., Nisar, A., & Kaushik, B. K. (2024) · Neuromorphic Computing and Engineering · doi:10.1088/2634-4386/ad2afa · Cited by 17
“The weight matrix W M × N from software training is with high precision as shown in figure 13(a) which used to generate a dictionary of K values using clustering function C(k) to replace the elements of W [28].”
MethodEnergy-Accuracy Trade-Offs for Resistive In-Memory Computing Architectures
Roy, S. K., & Shanbhag, N. R. (2024) · IEEE Journal on Exploratory Solid-State Computational Devices and Circuits · doi:10.1109/jxcdc.2024.3381888 · Cited by 16
“Recently, resistive parallel-bar IMCs have become the de-facto standard when designing complete large-scale IMC systems [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14] due to the ease of scalability afforded by the intrinsic isolation between ADC columns.”
BackgroundMethodInfluentialCherupally, S. K., Meng, J., Rakin, A. S., Yin, S., Yeo, I., Yu, S., Fan, D., & Seo, J.-S. (2021) · Semiconductor Science and Technology · doi:10.1088/1361-6641/ac461f · Cited by 12
“Recent IMC prototype chips have reported high energy efficiency [7, 9–12].”
InfluentialSelf-citationA Mini Tutorial of Processing in Memory: From Principles, Devices to Prototypes
Pan, B., Wang, G., Zhang, H., Kang, W., & Zhao, W. S. (2022) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2022.3172494 · Cited by 15
Ko, J., Kwon, D., Hwang, J., Lee, K., Oh, S., Kim, J., Im, J., Koo, R., Kim, J., & Lee, J. (2024) · Advanced Intelligent Systems · doi:10.1002/aisy.202300456 · Cited by 14
Compute SNDR-Boosted 22-nm MRAM-Based In-Memory Computing Macro Using Statistical Error Compensation
Roy, S. K., Ou, H.-M., Ahmed, M. G., Deaville, P., Zhang, B., Verma, N., Hanumolu, P. K., & Shanbhag, N. R. (2024) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2024.3442013 · Cited by 13
“As a result, the high-sensitivity readout peripherals in eNVM IMCs today occupy > 90% of the total IMC macro area [9], [10], [11], [15] thereby negating the areal density advantages of eNVM devices.”
Deaville, P., Zhang, B., & Verma, N. (2024) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2024.3512360 · Cited by 13
“For this reason, several previous eNVM IMC works have focused on RRAM, based either on high resistance-contrast technologies not available in typical commercial foundries or limited to preforming states of the RRAM [19], [20], [21], [22], both of which limit applicability.”
InfluentialXie, T., Yu, S., & Li, S. (2023) · IEEE Journal on Exploratory Solid-State Computational Devices and Circuits · doi:10.1109/jxcdc.2023.3255788 · Cited by 15
“…the sneak-path current issue and maintain better retention/reliability, practical RRAM-CIM macros are commonly implemented using a 1T1R array with binary cellwise computation [10], [11], [12], [13], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], as illustrated in Fig.”
BackgroundMethodResultInfluentialSelf-citationCharacterization and Mitigation of Relaxation Effects on Multi-level RRAM based In-Memory Computing
He, W., Shim, W., Yin, S., Sun, X., Fan, D., Yu, S., & Seo, J.-S. (2021) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps46558.2021.9405228 · Cited by 15
“Recent array-level demonstrations have presented RRAM’s potential towards IMC for area-/energy-efficient DNN inference [3-7], but most RRAM based IMC prototypes today feature only single-level cell design [3-5].”
MethodSelf-citationKoh, E. K., Dananjaya, P. A., Liu, L., Lee, C. X. X., Lim, G. J., You, Y. S., & Lew, W. S. (2024) · ACS Nano · doi:10.1021/acsnano.4c07454 · Cited by 12
Caselli, M., Debacker, P., & Boni, A. (2022) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2022.3174622 · Cited by 10
Neural Network Training Acceleration With RRAM-Based Hybrid Synapses
Choi, W., Kwak, M., Kim, S., & Hwang, H. (2021) · Frontiers in Neuroscience · doi:10.3389/fnins.2021.690418 · Cited by 13
Sparse and Robust RRAM-based Efficient In-memory Computing for DNN Inference
Meng, J., Yeo, I., Shim, W., Yang, L., Fan, D., Yu, S., & Seo, J.-S. (2022) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps48227.2022.9764480 · Cited by 10
“In this work, we obtained actual RRAM conductance variation across different temperatures from our RRAM prototype chip [15].”
MethodSelf-citationAdvancing Toward 4F² 1T1R RRAM With Local NAND-gate and Isolation Scheme
Bao, S., Wang, Z., Yang, Y., Wang, Q., Shan, L., Bao, L., Cai, Y., & Huang, R. (2025) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2025.3554165 · Cited by 10
“The limit for 1T1R cell area is 6F 2 , whereas RRAM achieves a minimal cell area of 4F 2 . is well compatible with CMOS back-end processes, which significantly extends the technology scalability across various technology nodes, from 130-to 12-nm node [10], [11], [12], [13], [14], [15], [16], [17].”
Ko, J., Im, J., Kim, J., Shin, W., Koo, R., Park, S., Woo, S. Y., & Lee, J. (2025) · Science Advances · doi:10.1126/sciadv.adt8227 · Cited by 9
Deaville, P., Zhang, B., Chen, L.-Y., & Verma, N. (2021) · European Solid-State Device Research Conference · doi:10.1109/essderc53440.2021.9631798 · Cited by 8
“While recent IMCs works have targeted ReRAM-based eNVM for row-parallel operation [1-4], IMC IP is required in the range of technologies suited for different application needs.”
BackgroundLi, Y., Wang, W. J., Wang, M., Dou, C., Ma, Z., Zhou, H., Zhang, P., Lepri, N., Zhang, X., Luo, Q., et al. (2023) · Advanced Intelligent Systems · doi:10.1002/aisy.202300399 · Cited by 8
Perspective: Entropy-stabilized oxide memristors
Chae, S., Yoo, S., Kioupakis, E., Lu, W. D., & Heron, J. T. (2024) · Applied Physics Letters · doi:10.1063/5.0216429 · Cited by 7
Strategies of high-accuracy memristor-based analogue computing in memory for artificial intelligence
Jiang, Z., Zhao, H., Tang, J., Lu, Y., Qin, Q., Wang, Z., Hu, R., Yu, R., He, Y., Zhang, J., et al. (2026) · Nature Materials · doi:10.1038/s41563-026-02600-y · Cited by 12
Adaptive Block Error Correction for Memristive Crossbars
Hemaram, S., Mayahinia, M., & Tahoori, M. B. (2022) · IEEE International On-Line Testing Symposium (IOLTS) · doi:10.1109/iolts56730.2022.9897817 · Cited by 7
Optimal Weight‐Splitting in Resistive Random Access Memory‐Based Computing‐in‐Memory Macros
Song, C., Kim, J., & Jeong, D. S. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200289 · Cited by 7
HfOx/Ge RRAM with High ON/OFF Ratio and Good Endurance
Wei, N., Ding, X. Y., Gao, S., Wu, W., & Zhao, Y. (2022) · Electronics · doi:10.3390/electronics11223820 · Cited by 7
Towards Efficient RRAM-based Quantized Neural Networks Hardware: State-of-the-art and Open Issues
Krestinskaya, O., Zhang, L., & Salama, K. N. (2022) · IEEE International Conference on Nanotechnology (IEEE-NANO) · doi:10.1109/nano54668.2022.9928590 · Cited by 7
“In [19], a fabricated BNN prototype (crossbar and peripheral readouts) is implemented with flash ADC, which is not practically scalable for higher bit precision networks.”
MethodInfluentialVertical AND-Type Flash Synaptic Cell Stack for High-Density and Reliable Binary Neural Networks
Kim, J., Im, J., Oh, S., Shin, W., Jung, G., Lee, S., & Lee, J. (2024) · IEEE Electron Device Letters · doi:10.1109/led.2024.3401399 · Cited by 8
“These advantages have led to active research on various synaptic devices, such as resistive random-access memory (RRAM) [6], [7], [8], magnetic RAM (MRAM) [9], and flash memory [10], [11], [12] for hardware-based BNNs.”
BackgroundXiao, S., Li, F., Hao, T., Xiao, L., Xiao, M., Mao, W., & Han, G. (2025) · Integrated Circuits and Systems · doi:10.23919/ics.2025.3571019 · Cited by 7
Demonstration of transformer-based ALBERT model on a 14nm analog AI inference chip
Chen, A., Ambrogio, S., Narayanan, P., Okazaki, A., Mackin, C., Fasoli, A., Rasch, M. J., Friz, A. M., Luquin, J., Yasuda, T., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-63794-4 · Cited by 13
“Figure 3f shows the hardware accuracy after layers 6, 9, 10, 11, and 12.”
ADC-Free ReRAM-Based In-Situ Accelerator for Energy-Efficient Binary Neural Networks
Kim, H., Jung, Y., & Kim, L. (2022) · IEEE Transactions on Computers · doi:10.1109/tc.2022.3224800 · Cited by 6
“Therefore, previous works [19], [21] adopted a complementary cell and word-line structure and assigned two adjacent cells in the bit-line direction for each weight element.”
BackgroundMethodInfluentialImplementation of an On-Chip Learning Neural Network IC Using Highly Linear Charge Trap Device
Choi, J.-M., Kwon, D.-W., Woo, J.-J., Park, E.-J., & Kwon, K.-W. (2021) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3071872 · Cited by 7
“The most recent Resistive RAM-based (RRAM) neural network IC has only implemented off-chip learning to represent edge computing [7]–[10].”
BackgroundPark, D., Park, H., Yeo, I., Kim, J., Lee, S., Lee, S., Shin, H., Pae, S.-I., Fan, D., Kung, J., et al. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3633487 · Cited by 11
“Yin et al. [141] fabricated Sun et al.’s [10] XNOR -RRAM using a 90-nm process and designed the peripheral circuits required for MAC operations.”
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-citationHeterogeneous Integration of In-Memory Analog Computing Architectures with Tensor Processing Units
Elbtity, M., Reidy, B., Amin, M. H., & Zand, R. (2023) · ACM Great Lakes Symposium on VLSI (GLSVLSI) · doi:10.1145/3583781.3590256 · Cited by 11
“Conventional IMC architectures typically employ emerging technologies such as resistive random access memory (RRAM) [6, 7] and magnetoresistive random-access memory (MRAM) [8] to accelerate matrix-vector multiplication (MVM) operations in the ML workloads through massive parallelism and analog computation.”
BackgroundInfluential
Metadata
- Authors
- Yin, S., Sun, X., Yu, S., & Seo, J.-S.
- Affiliations
- Arizona State University (US)
- Georgia Institute of Technology (US)
- Venue
- IEEE Transactions on Electron Devices
- Published
- 2020-10
- Volume
- 67
- Issue
- 10
- Pages
- 4185-4192
- ISSN
- 0018-9383 (ISSN-L)
- 1557-9646
- OpenAlex
- W2974585810
- Cited by
- 171
- References
- 50
- Citations by year
- 2026: 8 · 2025: 26 · 2024: 35 · 2023: 30 · 2022: 38 · 2021: 24 · 2020: 9
- FWCI
- 8.68
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Electron Devices (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- NSF-SRC-E2CDA (2018-NC-2762B)
- NSF (1652866, 1715443, 1740225)
- JUMP C-BRIC
- JUMP ASCENT (SRC Program sponsored by the Defense Advanced Research Projects Agency
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-power high-performance VLSI design
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Ferroelectric and Negative Capacitance Devices
- Keywords
- in-memory computing, resistive random access memory, binary neural networks, monolithic integration, 90 nm CMOS, analog-to-digital converter, energy-delay product, edge computing, neuromorphic hardware, MNIST, CIFAR-10
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.