Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks
The DOI matches its Crossref record; 38 of the 42 references it cites were found in a registry or an index.
Abstract
Eyeriss is an accelerator for state-of-the-art deep convolutional neural networks (CNNs). It optimizes for the energy efficiency of the entire system, including the accelerator chip and off-chip DRAM, for various CNN shapes by reconfiguring the architecture. CNNs are widely used in modern AI systems but also bring challenges on throughput and energy efficiency to the underlying hardware. This is because its computation requires a large amount of data, creating significant data movement from on-chip and off-chip that is more energy-consuming than computation. Minimizing data movement energy cost for any CNN shape, therefore, is the key to high throughput and energy efficiency. Eyeriss achieves these goals by using a proposed processing dataflow, called row stationary (RS), on a spatial architecture with 168 processing elements. RS dataflow reconfigures the computation mapping of a given shape, which optimizes energy efficiency by maximally reusing data locally to reduce expensive data movement, such as DRAM accesses. Compression and data gating are also applied to further improve energy efficiency. Eyeriss processes the convolutional layers at 35 frames/s and 0.0029 DRAM access/multiply and accumulation (MAC) for AlexNet at 278 mW (batch size N = 4), and 0.7 frames/s and 0.0035 DRAM access/MAC for VGG-16 at 236 mW (N = 3).
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
“such scaling will enable NeuRRAM to deliver higher energy and area efficiency than today’s state-of-the-art edge inference accelerators58–61.”
From A compute-in-memory chip based on resistive random-access memory (2022) · cites this paper as [58]
In the citing paper
“Some studies proposed to implement the argmax() function via hardware128–148, which could be beneficial to reduce the total transistor count and power consumption while at the same time increasing the throughput.”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [131]
In the citing paper
“Based on the analysis of the research, CoNNA is able to speed up at least 3.08 times and up to 14.1 times faster than the MIT Eyeriss Accelerator [20].”
From Embedded Intelligence on FPGA: Survey, Applications and Challenges (2021) · cites this paper as [20]
In the citing paper
“massively parallel processing architectures with distributed memory architectures, such as graphic processing units (GPUs) but also increasingly specialized chips for NN inference (e.g., [17]), have shown to be very energy efficient and capable of impressive computational throughput when fully utilized by concurrent algorithms [18].”
From An Introduction to Deep Learning for the Physical Layer (2017) · cites this paper as [17]
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×
“Modern CNNs achieve their superior accuracy by building a very deep hierarchy of layers [2]–[5], [7], which transform the input image data into highly abstract representations called feature maps (fmaps).”
BackgroundVery 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 7×
“We benchmark the chip performance using two publicly available and widely used CNNs: AlexNet [2] and VGG-16 [3].”
MethodInfluentialImageNet 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
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 1×
“[25] demonstrate a CNN processor and only report the theoretical peak throughput along with the power measured on a CNN for the MNIST data set [26], which has storage and computation requirements that are orders of magnitude lower than the state-of-the-art CNNs.”
MethodGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 2×
“We have demonstrated an 1000-class image classification task [27] using this system, and a live demo can be found in [29].”
BackgroundMethodRich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
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×
“D EEP learning using convolutional neural networks (CNNs) [1] has achieved unprecedented accuracy on many modern AI applications [2]–[9].”
Background(2010) · International Conference on Machine Learning · Cited by 13,995
Research works printed but not linked: 2
- han, 2015, Learning both weights and connections for efficient neural network, Proc Adv Neural Inf Process Syst, 28, 1135
- chen, 2016, An 1000-Class Image Classification Task Performed by the Eyeriss-Integrated Deep Learning System
Cited by
Items 401–500 of 1,001
Xuan, L., Un, K.-F., Lam, C., & Martins, R. P. (2022) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2022.3180553 · Cited by 66
“contributed to the design of energy-efficient DNN accelerators for graphics processing units (GPUs) [2], application-specific integrated circuits (ASICs) [3]–[7], and field-programmable”
BackgroundA write‐improved low‐power 12T SRAM cell for wearable wireless sensor nodes
Sharma, V., Vishvakarma, S. K., Chouhan, S. S., & Halonen, K. A. I. (2018) · International Journal of Circuit Theory and Applications · doi:10.1002/cta.2555 · Cited by 52
Giacomin, E., Greenberg-Toledo, T., Kvatinsky, S., & Gaillardon, P. (2018) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2018.2872455 · Cited by 53
Enabling Design Methodologies and Future Trends for Edge AI: Specialization and Codesign
Hao, C., Dotzel, J., Xiong, J., Benini, L., Zhang, Z., & Chen, D. (2021) · IEEE Design and Test · doi:10.1109/mdat.2021.3069952 · Cited by 67
Harnessing optoelectronic noises in a photonic generative network
Wu, C., Yang, X., Yu, H., Peng, R., Takeuchi, I., Chen, Y., & Li, M. (2022) · Science Advances · doi:10.1126/sciadv.abm2956 · Cited by 64
Self-citationOpenSMART: Single-cycle multi-hop NoC generator in BSV and Chisel
Kwon, H., & Krishna, T. (2017) · IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) · doi:10.1109/ispass.2017.7975291 · Cited by 51
“The domains vary from (i) many-core chips in HPC supercomputers and high-end servers with tens to hundreds of homogeneous cores [1], [2], [3], to (ii) mobile and embedded SoCs with tens of heterogeneous cores and controllers [4], [5], to (iii) GPUs with hundreds of SMs [6], to (iv) domain-specific accelerators, such a…”
BackgroundSelf-citationSpatial Correlation and Value Prediction in Convolutional Neural Networks
Shomron, G., & Weiser, U. C. (2018) · IEEE Computer Architecture Letters · doi:10.1109/lca.2018.2890236 · Cited by 56
“We observe that 73%, 60%, and 41% of activation are zero-valued for AlexNet, VGG-16 and ResNet-18, respectively [8, 9, 10].”
BackgroundDynExit: A Dynamic Early-Exit Strategy for Deep Residual Networks
Wang, M., Mo, J., Lin, J. T., Wang, Z., & Du, L. (2019) · SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems · doi:10.1109/sips47522.2019.9020551 · Cited by 68
Li, B., Wang, H., Zhang, X., Ren, J., Liu, L., Sun, H., & Zheng, N. (2021) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3078541 · Cited by 56
Eager pruning: algorithm and architecture support for fast training of deep neural networks
Zhang, J., Chen, X., Song, M., & Li, T. (2019) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3307650.3322263 · Cited by 80
An Approximate Memory Architecture for Energy Saving in Deep Learning Applications
Nguyen, D. T., Hung, N. H., Kim, H., & Lee, H. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2019.2962516 · Cited by 59
Parallel Time Batching: Systolic-Array Acceleration of Sparse Spiking Neural Computation
Lee, J.-J., Zhang, W., & Li, P. (2022) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca53966.2022.00031 · Cited by 73
Jain, V., Giraldo, J. S. P., Roose, J. D., Mei, L., Boons, B., & Verhelst, M. (2023) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2023.3236566 · Cited by 62
Lee, J., Mukhanov, L., Molahosseini, A. S., Minhas, U. I., Yang, H., del Rincón, J. M., Dichev, K., Hong, C.-H., & Vandierendonck, H. (2023) · ACM Computing Surveys · doi:10.1145/3587095 · Cited by 69
“Exploring binary weights [38] with binary inputs offered the opportunity to explore XNOR gates for the efficient implementation of CNN [123], improving the accuracy per memory foot print and per Joule.”
BackgroundNeuromorphic applications in medicine
Aboumerhi, K., Güemes, A., Liu, H., Tenore, F. V., & Etienne‐Cummings, R. (2023) · Journal of Neural Engineering · doi:10.1088/1741-2552/aceca3 · Cited by 59
“[87] Chen Y-H, Krishna T, Emer J S and Sze V 2017 Eyeriss: an”
Neuromorphic computing's yesterday, today, and tomorrow – an evolutional view
Chen, Y., Li, H. H., Wu, C., Song, C., Li, S. K., Min, C., Cheng, H.-P., Wen, W., & Liu, X. (2017) · Integration · doi:10.1016/j.vlsi.2017.11.001 · Cited by 51
Self-citationDouble MAC on a DSP: Boosting the Performance of Convolutional Neural Networks on FPGAs
Lee, S., Kim, D. W., Nguyen, D. T., & Lee, J. (2018) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2018.2824280 · Cited by 60
GNA: Reconfigurable and Efficient Architecture for Generative Network Acceleration
Yan, J., Yin, S., Tu, F., Liu, L., & Wei, S. (2018) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2018.2857258 · Cited by 54
“The traditional convolution neural network accelerators on embedded devices like Eyeriss [9], Envision [10], and more [11], they all have the basic components like 2-”
BackgroundAn MRAM-Based Deep In-Memory Architecture for Deep Neural Networks
Patil, A. D., Hua, H., Gonugondla, S., Kang, M., & Shanbhag, N. R. (2019) · 2019 IEEE International Symposium on Circuits and Systems (ISCAS) · doi:10.1109/iscas.2019.8702206 · Cited by 49
“Recently, near-memory and in-memory computing approaches [4]–[18] are being widely explored to achieve significant energy-delay benefits over conventional digital implementations [19], [20].”
BackgroundFPGA-based CNN inference accelerator synthesized from multi-threaded C software
Kim, J. H., Grady, B., Lian, R., Brothers, J., & Anderson, J. H. (2017) · IEEE International System-on-Chip Conference (SOCC) · doi:10.1109/socc.2017.8226056 · Cited by 58
Neural Architecture Search and Hardware Accelerator Co-Search: A Survey
Sekanina, L. (2021) · IEEE Access · doi:10.1109/access.2021.3126685 · Cited by 60
A Precision-Scalable Energy-Efficient Convolutional Neural Network Accelerator
Liu, W., Lin, J. T., & Wang, Z. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.2993051 · Cited by 58
Hammad, I., Li, L., El‐Sankary, K., & Snelgrove, W. M. (2021) · IEEE Access · doi:10.1109/access.2021.3049299 · Cited by 59
An Overview of Energy-Efficient Hardware Accelerators for On-Device Deep-Neural-Network Training
Lee, J., & Yoo, H. (2021) · IEEE Open Journal of the Solid-State Circuits Society · doi:10.1109/ojsscs.2021.3119554 · Cited by 64
FitNN: A Low-Resource FPGA-Based CNN Accelerator for Drones
Zhang, Z., Mahmud, M. A. P., & Kouzani, A. Z. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3179016 · Cited by 58
“Since transferring data between on-chip and offchip would cause relatively high latency and power consumption, it is necessary to avoid repeated data loading [20].”
BackgroundSTONNE: Enabling Cycle-Level Microarchitectural Simulation for DNN Inference Accelerators
Muñoz-Martínez, F., Abellán, J. L., Acacio, M. E., & Krishna, T. (2021) · IEEE International Symposium on Workload Characterization · doi:10.1109/iiswc53511.2021.00028 · Cited by 56
Self-citationDictionary Representation of Deep Features for Occlusion-Robust Face Recognition
Cen, F., & Wang, G. (2019) · IEEE Access · doi:10.1109/access.2019.2901376 · Cited by 51
“With the development of deep neural network accelerator chips [20], the CNN-based feature extraction will gradually shift to the terminal or mobile end.”
MethodShortcut Mining: Exploiting Cross-Layer Shortcut Reuse in DCNN Accelerators
Mazreah, A. A., & Chen, L. (2019) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca.2019.00030 · Cited by 51
“, [2][4][20]), recent advancement in DCNNs and hardware platforms have led to changed behaviors and properties of the network models, which results in new opportunities for data reusing.”
BackgroundInfluentialIm, D., Han, D., Choi, S., Kang, S., & Yoo, H. (2019) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas.2019.8702243 · Cited by 52
“Since both D-CONV and T-CONV can be replaced by the normal 2D convolution by inserting the virtual zeros in either kernel map or input feature map, previous CNN accelerators [3, 6, 7] can compute both D-CONV and TCONV by feeding the kernels and input feature map with the virtual zeros.”
MethodYang, C., Wang, Y., Wang, X., & Geng, L. (2019) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2019.2928682 · Cited by 49
“Unlike the ASIC based works [12], [13], [17], WRA supports popular CNN models, such as MobileNet, and still achieves 6× to 10.”
BackgroundMethodResultInfluentialHardware Implementation of Neural Self-Interference Cancellation
Kurzo, Y., Kristensen, A. T., Burg, A. P., & Balatsoukas‐Stimming, A. (2020) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2020.2992370 · Cited by 57
“Existing NN hardware accelerators, such as [26], [27], mainly target applications where both the size of the NN and the number of inputs is very large, and where producing a few tens of outputs per second is sufficient.”
BackgroundZhao, Y. K., Li, C., Wang, Y., Xu, P., Zhang, Y., & Lin, Y. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053977 · Cited by 60
“4: Comparison on the inference latency from Eyeriss [29] and our Predictor when running AlexNet.”
BackgroundChoi, S., Sim, J., Kang, M., Choi, Y., Kim, H., & Kim, L. (2020) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2020.3005786 · Cited by 54
DANCE: Differentiable Accelerator/Network Co-Exploration
Choi, K., Hong, D., Yoon, H., Yu, J., Kim, Y., & Lee, J. (2021) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1109/dac18074.2021.9586121 · Cited by 61
DrAcc: a DRAM based Accelerator for Accurate CNN Inference
Deng, Q., Jiang, L., Zhang, Y., Zhang, M., & Yang, J. (2018) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1109/dac.2018.8465866 · Cited by 73
Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural NetworksPreprint
Sharma, H., Park, J., Suda, N., Lai, L., Chau, B., Kim, J. K., Chandra, V., & Esmaeilzadeh, H. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.01507 · Cited by 49
“Eyeriss [1, 3] presents an optimized row-stationary dataflow for DNNs to improve efficiency.”
BackgroundInfluentialGradPIM: A Practical Processing-in-DRAM Architecture for Gradient Descent
Kim, H., Park, H., Kim, T., Cho, K., Lee, E., Ryu, S., Lee, H., Choi, K., & Lee, J. (2021) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca51647.2021.00030 · Cited by 65
“This is analogous to the weight-stationary dataflow [23], [41] often used in systolic array architectures [27], [36], [49].”
BackgroundMethodHardware-aware machine learning: modeling and optimization
Marculescu, D., Stamoulis, D., & Cai, E. (2018) · Proceedings of the International Conference on Computer-Aided Design · doi:10.1145/3240765.3243479 · Cited by 49
“As opposed to Eyeriss, Paleo has been developed with general purpose platforms (GPUs) in mind.”
BackgroundMethodInfluentialSensitivity based Error Resilient Techniques for Energy Efficient Deep Neural Network Accelerators
Choi, W., Shin, D., Park, J., & Ghosh, S. (2019) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3316781.3317908 · Cited by 48
“Recently, various energy efficient DNN accelerators using application specific integrated circuit (ASIC) have been proposed [4].”
BackgroundMASR: A Modular Accelerator for Sparse RNNs
Gupta, U., Reagen, B., Pentecost, L., Donato, M., Tambe, T., Rush, A. M., Wei, G.-Y., & Brooks, D. (2019) · 2019 28th International Conference on Parallel Architectures and Compilation Techniques (PACT) · doi:10.1109/pact.2019.00009 · Cited by 47
“However, much of the effort has been devoted to optimizing DNNs with only FC and CNN layers [8], [9], [10], [11], [13], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34].”
BackgroundToward Full-Stack Acceleration of Deep Convolutional Neural Networks on FPGAs
Liu, S., Fan, H., Ferianc, M., Niu, X., Shi, H.-F., & Luk, W. W. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3055240 · Cited by 57
Albireo: Energy-Efficient Acceleration of Convolutional Neural Networks via Silicon Photonics
Shiflett, K., Karanth, A., Bunescu, R., & Louri, A. (2021) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1109/isca52012.2021.00072 · Cited by 66
A Low-Power Deep Neural Network Online Learning Processor for Real-Time Object Tracking Application
Han, D., Lee, J., Lee, J., & Yoo, H. (2018) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2018.2880363 · Cited by 48
“We reduce on-chip memory size by using run-length compression (RLC) which is introduced in previous DNN inference accelerator [13].”
MethodCompiling Spiking Neural Networks to Neuromorphic Hardware
Song, S., Balaji, A., Das, A. K., Kandasamy, N., & Shackleford, J. (2020) · doi:10.1145/3372799.3394364 · Cited by 54
Nonconventional Computer Arithmetic Circuits, Systems and Applications
Sousa, L. A. (2021) · IEEE Circuits and Systems Magazine · doi:10.1109/mcas.2020.3027425 · Cited by 53
“The fully RNS-based CNN accelerator in [123], i.e., Res-DNN , is supported by the Eyeriss architecture [124] and the RNS arithmetic for the 3-moduli set , , , 2 1 2 2 1 n n - , preventing overflow by using the BE algorithm referenced in II-B (extending 2 n to the modulo ).”
MethodCaffePresso: an optimized library for deep learning on embedded accelerator-based platforms
Hegde, G., Siddhartha, Ramasamy, N., & Kapre, N. (2016) · Proceedings of the International Conference on Compilers, Architectures and Synthesis for Embedded Systems · doi:10.1145/2968455.2968511 · Cited by 46
“While there is significant interest in design of ASICs customized for deep learning [5, 8, 6, 2, 3], along with large FPGA-based [10, 14, 7] and GPU-based accelerators [13] for high-performance systems, we investigate the potential of commercial off-the-shelf SoC hardware for efficiently implementing these tasks at lo…”
BackgroundRubik: A Hierarchical Architecture for Efficient Graph Neural Network Training
Chen, X., Wang, Y., Xie, X., Hu, X., Basak, A., Liang, L., Yan, M., Deng, L., Ding, Y., Du, Z., et al. (2021) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2021.3079142 · Cited by 72
From Near-Sensor to In-Sensor: A State-of-the-Art Review of Embedded AI Vision Systems
Fabre, W., Haroun, K., Lorrain, V., Lepecq, M., & Sicard, G. (2024) · Sensors · doi:10.3390/s24165446 · Cited by 63
“To overcome this challenge, some approaches use data-reorganization methods, such as compressed sparse column (CSC) [117], to efficiently manage and access sparse information.”
MethodPIR-DSP: An FPGA DSP Block Architecture for Multi-precision Deep Neural Networks
Rasoulinezhad, S., Zhou, H., Wang, L., & Leong, P. H. W. (2019) · IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM) · doi:10.1109/fccm.2019.00015 · Cited by 60
“Low energy and high performance DNN accelerators have been demonstrated using systolic array architectures [9], [10].”
BackgroundMethodInfluentialYang, C., Wang, Y., Wang, X., & Geng, L. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.2985727 · Cited by 46
RAPIDNN: In-Memory Deep Neural Network Acceleration FrameworkPreprint
Imani, M., Samragh, M., Kim, Y., Gupta, S., Koushanfar, F., & Rosing, T. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.05794 · Cited by 59
Bavandpour, M., Sahay, S., Mahmoodi, M. R., & Strukov, D. B. (2021) · Neuromorphic Computing and Engineering · doi:10.1088/2634-4386/ac0775 · Cited by 45
FPGA Based Implementations of RNN and CNN: A Brief Analysis
Mahreen, Z., Usmani, A. R., Mehrban, S., & Hussain, M. (2019) · International Conference on Innovative Computing (ICIC) · doi:10.1109/icic48496.2019.8966676 · Cited by 47
Kim, S., Kim, S., Cho, K., Shin, T., Park, H., Lho, D., Park, S., Son, K., Park, G., Jeong, S., et al. (2021) · IEEE Transactions on Components Packaging and Manufacturing Technology · doi:10.1109/tcpmt.2021.3117071 · Cited by 54
LISA: Graph Neural Network based Portable Mapping on Spatial Accelerators
Li, Z., Wu, D., Wijerathne, D., & Mitra, T. (2022) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca53966.2022.00040 · Cited by 49
Cerebron: A Reconfigurable Architecture for Spatiotemporal Sparse Spiking Neural Networks
Chen, Q., Gao, C., & Fu, Y. (2022) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2022.3196839 · Cited by 58
Neuromorphic Photonics Circuits: Contemporary Review
Kutluyarov, R. V., Zakoyan, A. G., Voronkov, G. S., Grakhova, E. P., & Butt, M. A. (2023) · Nanomaterials · doi:10.3390/nano13243139 · Cited by 53
Omid-Zohoor, A., Young, C. N., Ta, D., & Murmann, B. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2653187 · Cited by 53
Reliability perspective of resistive synaptic devices on the neuromorphic system performance
Chen, P.-Y., & Yu, S. (2018) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps.2018.8353615 · Cited by 48
“VI. REFERENCE [1] Y.-H. Chen, et al., “Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks,” IEEE International Solid-State Circuits Conference (ISSCC), pp. 262-263, 2016.”
MethodInfluentialJoint Feature and Texture Coding: Toward Smart Video Representation via Front-End Intelligence
Ma, S., Zhang, X., Wang, S., Zhang, X., Jia, C., & Wang, S. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2873102 · Cited by 52
Amaravati, A., Nasir, S. B., Ting, J., Yoon, I., & Raychowdhury, A. (2018) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2018.2881288 · Cited by 45
Resilient Low Voltage Accelerators for High Energy Efficiency
Chandramoorthy, N., Swaminathan, K., Cochet, M., Paidimarri, A., Eldridge, S., Joshi, R., Ziegler, M. M., Buyuktosunoglu, A., & Bose, P. K. (2019) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca.2019.00034 · Cited by 64
“Several prior works such as [18, 25, 26, 27, 28, 29] showcase custom hardware designs for cognitive applications.”
BackgroundMethodCai, R., Ren, A., Chen, O., Liu, N., Ding, C., Qian, X., Han, J., Luo, W., Yoshikawa, N., & Wang, Y. (2019) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3307650.3322270 · Cited by 56
“To simultaneously achieve high performance and energy efficiency, hardware acceleration of DNNs , including FPGA-and ASIC-based implementations, has been extensively investigated [4, 6–9, 12, 13, 17–20, 28, 29, 33, 34, 37, 39– 41, 48, 49, 51, 53–55].”
MethodIm, D., Han, D., Choi, S., Kang, S., & Yoo, H. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.2991189 · Cited by 51
“The previous work [7] accelerated CNNs by integrating a large number of processing elements (PEs) with on-chip data reuse.”
BackgroundMethodSWM: A High-Performance Sparse-Winograd Matrix Multiplication CNN Accelerator
Wu, D., Fan, X., Cao, W., & Wang, L. (2021) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2021.3060041 · Cited by 45
Efficient Processing of Deep Neural Networks: A Tutorial and SurveyPreprint
Sze, V., Chen, Y., Yang, T.-J., & Emer, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.09039 · Cited by 45
Self-citationSmartExchange: Trading Higher-cost Memory Storage/Access for Lower-cost Computation
Zhao, Y. K., Chen, X., Wang, Y., Li, C., You, H., Fu, Y., Xie, Y., Wang, Z., & Lin, Y. (2020) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1109/isca45697.2020.00082 · Cited by 53
Wang, P., Ji, Y., Hong, C., Lyu, Y., Wang, D., & Xie, Y. (2018) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3195970.3196116 · Cited by 94
Lym, S., Lee, D., O’Connor, M., Chatterjee, N., & Erez, M. (2019) · IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) · doi:10.1109/ispass.2019.00041 · Cited by 63
Wang, Y., Ding, C., Li, Z., Yuan, G., Liao, S., Ma, X., Yuan, B., Qian, X., Tang, J., Qiu, Q., et al. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11653 · Cited by 44
A Raw Image-Based End-to-End Object Detection Accelerator Using HOG Features
Zhang, X., Zhang, L., & Lou, X. (2021) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3098053 · Cited by 44
Garofalo, A., Ottavi, G., Conti, F., Karunaratne, G., Boybat, I., Benini, L., & Rossi, D. (2022) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2022.3170152 · Cited by 49
Zhao, Z., Cao, R., Un, K.-F., Yu, W.-H., Mak, P., & Martins, R. P. (2022) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2022.3196055 · Cited by 56
“than the application-specific integrated circuits [1]–[4] and is more cost-effective than the GPU solutions [5].”
BackgroundResistive-RAM-Based In-Memory Computing for Neural Network: A Review
Chen, W., Qi, Z., Akhtar, Z., & Siddique, K. (2022) · Electronics · doi:10.3390/electronics11223667 · Cited by 51
Lin, C.-T., Wang, D., Zhang, B., Chen, G. K., Knag, P. C., Krishnamurthy, R. K., & Seok, M. (2023) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2023.3313519 · Cited by 49
“To mitigate the complexity, recent works have proposed specialized hardware architectures [6], [7], [8], [9], [10].”
BackgroundKurzo, Y., Burg, A. P., & Balatsoukas‐Stimming, A. (2018) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/acssc.2018.8645295 · Cited by 51
“Existing NN hardware accelerators, such as [10], [11], mainly target applications where both the size of the NN and the number of inputs is very large, and where producing a a few tens of outputs per second is sufficient.”
BackgroundSkippyNN: An Embedded Stochastic-Computing Accelerator for Convolutional Neural Networks
Hojabr, R., Givaki, K., Tayaranian, S. M. R., Esfahanian, P., Khonsari, A., Rahmati, D., & Najafi, M. H. (2019) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3316781.3317911 · Cited by 60
“Customized hardware implementations have gained a lot of attention in recent years to tackle these challenges [5, 6, 25].”
BackgroundMethodInfluentialVWA: Hardware Efficient Vectorwise Accelerator for Convolutional Neural Network
Chang, K.-W., & Chang, T. (2019) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2019.2942529 · Cited by 48
Automatic generation of high-performance quantized machine learning kernels
Cowan, M., Moreau, T., Chen, T., Bornholt, J., & Ceze, L. (2020) · Proceedings of the CGO/Proceedings of the ... CGO · doi:10.1145/3368826.3377912 · Cited by 55
“The emergence of machine learning accelerators in ASICs [6, 7, 19] and FPGAs [10, 25, 41] has led to increased specialization of both hardware intrinsics (e.g. single instruction matrix multiplication) and data type selection.”
BackgroundEnergy-efficient deep learning inference on edge devices
Daghero, F., Pagliari, D. J., & Poncino, M. (2020) · Advances in computers · doi:10.1016/bs.adcom.2020.07.002 · Cited by 49
Alioto, M., De, V. K., & Marongiu, A. (2018) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2018.2881461 · Cited by 49
MnnFast: a fast and scalable system architecture for memory-augmented neural networks
Jang, H., Kim, J., Jo, J.-E., Lee, J., & Kim, J. (2019) · Proceedings of the 46th International Symposium on Computer Architecture · doi:10.1145/3307650.3322214 · Cited by 42
“[12] classify dataflows into various types, and propose a row stationary type accelera-”
BackgroundCommunication Lower Bound in Convolution Accelerators
Chen, X., Han, Y., & Wang, Y. (2020) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca47549.2020.00050 · Cited by 49
“[10] reported the DRAM access volume of VGGNet-16 with input compression on Eyeriss.”
BackgroundMethodResultInfluentialIntermittent-Aware Neural Architecture Search
Mendis, H. R., Kang, C.-K., & Hsiu, P.-C. (2021) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3476995 · Cited by 52
“As ambient power is typically unstable and too weak for continuous execution [47], even extremely power-efficient hardware [10, 21], may take multiple power-cycles to complete a full inference.”
BackgroundSparsely-Connected Neural Networks: Towards Efficient VLSI Implementation of Deep Neural NetworksPreprint
Ardakani, A., Condo, C., & Gross, W. J. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.01427 · Cited by 74
“Due to parameter reuse in convolutional layers, they are well-studied and can be efficiently implemented with customized hardware platforms (Chen et al. (2016); Shafiee et al. (2016a); Chen et al. (2016)).”
Method10T SRAM Computing-in-Memory Macros for Binary and Multibit MAC Operation of DNN Edge Processors
Nguyen, V. T., Kim, J.-S., & Lee, J. (2021) · IEEE Access · doi:10.1109/access.2021.3079425 · Cited by 43
Wu, N. (2018) · Science China Information Sciences · doi:10.1007/s11432-017-9303-0 · Cited by 41
Neuromorphic LIF Row-by-Row Multi-convolution Processor for FPGA
Tapiador-Morales, R., Linares-Barranco, A., Jimenez-Fernandez, A. F., & Jiménez, G. (2018) · IEEE Transactions on Biomedical Circuits and Systems · doi:10.1109/tbcas.2018.2880012 · Cited by 45
NAND-Net: Minimizing Computational Complexity of In-Memory Processing for Binary Neural Networks
Kim, H., Sim, J., Choi, Y., & Kim, L. (2019) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca.2019.00017 · Cited by 51
Bit Prudent In-Cache Acceleration of Deep Convolutional Neural Networks
Wang, X., Yu, J., Augustine, C., Iyer, R., & Das, R. (2019) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca.2019.00029 · Cited by 46
“Eyeriss [24] is a low-power DNN accelerator.”
BackgroundPre-defined Sparsity for Low-Complexity Convolutional Neural Networks
Kundu, S., Nazemi, M., Pedram, M., Chugg, K. M., & Beerel, P. A. (2020) · IEEE Transactions on Computers · doi:10.1109/tc.2020.2972520 · Cited by 51
“In particular, as implemented in Eyeriss [13], at 62% density, RLC would lead to 0.”
BackgroundMethodKlessydra-T: Designing Vector Coprocessors for Multithreaded Edge-Computing Cores
Cheikh, A., Sordillo, S., Mastrandrea, A., Menichelli, F., Scotti, G., & Olivieri, M. (2021) · IEEE Micro · doi:10.1109/mm.2021.3050962 · Cited by 51
“Also in [10,11], energy efficient accelerators were designed to be efficient by making highly parallel hardware units and minimizing off-chip data movements by reusing on-chip loaded data.”
MethodTensor Casting: Co-Designing Algorithm-Architecture for Personalized Recommendation Training
Kwon, Y., Lee, Y., & Rhu, M. (2021) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca51647.2021.00029 · Cited by 55
Deep In-Memory Architectures in SRAM: An Analog Approach to Approximate Computing
Kang, M., Gonugondla, S. K., & Shanbhag, N. R. (2020) · Proceedings of the IEEE · doi:10.1109/jproc.2020.3034117 · Cited by 41
“These trends provide another justification for the use of DIMA to realize DNNs since it has been shown that the analog computations are more energy-efficient than the digital at lower SNRs (SNR 60 dB) [86] and because the deeper layers tend to be memory bound as the opportunities for weight reuse decrease.”
BackgroundMethodInfluentialAtomic Dataflow based Graph-Level Workload Orchestration for Scalable DNN Accelerators
Zheng, S., Zhang, X., Liu, L., Wei, S., & Yin, S. (2022) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca53966.2022.00042 · Cited by 58
Moitra, A., Bhattacharjee, A., Kuang, R., Krishnan, G., Cao, Y. K., & Panda, P. (2023) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2023.3274918 · Cited by 56
HighLight: Efficient and Flexible DNN Acceleration with Hierarchical Structured Sparsity
Wu, Y. N., Tsai, P.-A., Muralidharan, S., Parashar, A., Sze, V., & Emer, J. (2023) · IEEE/ACM International Symposium on Microarchitecture (MICRO) · doi:10.1145/3613424.3623786 · Cited by 48
Self-citationSurvey on Activation Functions for Optical Neural Networks
Destras, O., Beux, S. L., de Magalhães, F. G., & Nicolescu, G. (2023) · ACM Computing Surveys · doi:10.1145/3607533 · Cited by 49
“recognition may have to implement billions of operations just for the processing of one image [2].”
BackgroundSPARCNet: A Hardware Accelerator for Efficient Deployment of Sparse Convolutional Networks
Page, A., Jafari, A., Shea, C., & Mohsenin, T. (2017) · ACM Journal on Emerging Technologies in Computing Systems · doi:10.1145/3005448 · Cited by 43
A survey of neural network accelerators
Li, Z., Wang, Y., Zhi, T., & Chen, T. (2017) · Frontiers of Computer Science · doi:10.1007/s11704-016-6159-1 · Cited by 46
“[80] which possesses a novel organization of computation units, memory hierarchy and on-chip network to exploit data reuse for different CNN shapes achieved 34.”
Background
Metadata
- Authors
- Chen, Y.-H., Krishna, T., Emer, J. S., & Sze, V.
- Affiliations
- Massachusetts Institute of Technology (US)
- Georgia Institute of Technology (US)
- Nvidia (United States) (US)
- Venue
- IEEE Journal of Solid-State Circuits
- Published
- 2017-01
- Volume
- 52
- Issue
- 1
- Pages
- 127-138
- ISSN
- 0018-9200 (ISSN-L)
- 1558-173X
- OpenAlex
- W2289252105
- Semantic Scholar
- ffdaa12ef011de9dbf43be46d45a3abcc8288965
- Cited by
- 3,247
- References
- 42
- Citations by year
- 2026: 218 · 2025: 313 · 2024: 294 · 2023: 373 · 2022: 365 · 2021: 450 · 2020: 427 · 2019: 354 · 2018: 302 · 2017: 118
- FWCI
- 182.12
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Journal of Solid-State Circuits (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Keywords
- reconfigurable accelerators, deep convolutional neural networks, energy efficiency, spatial architecture, data movement reduction, data compression, processing elements, AlexNet, VGG16, convolutional layers
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.