Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks
The DOI matches its Crossref record; 17 of the 17 references it cites were found in a registry or an index.
Abstract
Convolutional neural network (CNN) has been widely employed for image recognition because it can achieve high accuracy by emulating behavior of optic nerves in living creatures. Recently, rapid growth of modern applications based on deep learning algorithms has further improved research and implementations. Especially, various accelerators for deep CNN have been proposed based on FPGA platform because it has advantages of high performance, reconfigurability, and fast development round, etc. Although current FPGA accelerators have demonstrated better performance over generic processors, the accelerator design space has not been well exploited. One critical problem is that the computation throughput may not well match the memory bandwidth provided an FPGA platform. Consequently, existing approaches cannot achieve best performance due to under-utilization of either logic resource or memory bandwidth. At the same time, the increasing complexity and scalability of deep learning applications aggravate this problem. In order to overcome this problem, we propose an analytical design scheme using the roofline model. For any solution of a CNN design, we quantitatively analyze its computing throughput and required memory bandwidth using various optimization techniques, such as loop tiling and transformation. Then, with the help of rooine model, we can identify the solution with best performance and lowest FPGA resource requirement. As a case study, we implement a CNN accelerator on a VC707 FPGA board and compare it to previous approaches. Our implementation achieves a peak performance of 61.62 GFLOPS under 100MHz working frequency, which outperform previous approaches significantly.
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
“Another option is to use field programable gate arrays (FPGAs), which consume much less energy than GPUs while providing an intermediate computing efficiency between CPUs and GPUs13–17.”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [15]
Citations
Cites
ImageNet 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 3×
“Figure 2 shows a real-life CNN application, taken from [9].”
BackgroundMethodGradient-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 algorithm receives three 224x224 input images that are from an original 256x256 three-channel RGB image.”
Advances in Neural Information Processing Systems 25
Zhang, Y., Sutton, C. A., Storkey, A., & Zoubin Ghahramani, A. (2012) · Cited by 12,653
3D Convolutional Neural Networks for Human Action Recognition
Ji, S., Xu, W., Yang, M., & Yu, K. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2012.59 · Cited by 6,235
Williams, S., Waterman, A., & Patterson, D. A. (2009) · Communications of the ACM · doi:10.1145/1498765.1498785 · Cited by 2,798
DianNao: a small-footprint high-throughput accelerator for ubiquitous machine-learning
Chen, T., Du, Z., Sun, N., Wang, J., Wu, C., Chen, Y., & Temam, O. (2014) · Proceedings of the 19th international conference on Architectural support for programming languages and operating systems · doi:10.1145/2541940.2541967 · Cited by 1,297
An empirical evaluation of deep architectures on problems with many factors of variation
Larochelle, H., Erhan, D., Courville, A., Bergstra, J., & Bengio, Y. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273556 · Cited by 992 · cited 1×
“Convolutional neural network (CNN), a well-known deep learning architecture extended from artificial neural network, has been extensively adopted in various applications, which include video surveillance, mobile robot vision, image search engine in data centers, etc [6] [7] [8] [10] [14].”
BackgroundMinimizing Computation in Convolutional Neural Networks
Cong, J., & Xiao, B. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-11179-7_36 · Cited by 372 · cited 1×
“In the feedforward computation perspective, a previous study [5] proved that convolution operations will occupy over 90% of the computation time.”
BackgroundSelf-citationCNP: An FPGA-based processor for Convolutional Networks
Farabet, C., Poulet, C., Han, J. Y., & LeCun, Y. (2009) · 2009 International Conference on Field Programmable Logic and Applications · doi:10.1109/fpl.2009.5272559 · Cited by 367
A dynamically configurable coprocessor for convolutional neural networks
Chakradhar, S., Sankaradas, M., Jakkula, V., & Cadambi, S. (2010) · Proceedings of the 37th annual international symposium on Computer architecture · doi:10.1145/1815961.1815993 · Cited by 343 · cited 8×
“Among these approaches, FPGA based accelerators have attracted more and more attention of researchers because they have advantages of good performance, high energy efficiency, fast development round, and capability of reconfiguration [1] [2] [3] [6] [12] [14].”
BackgroundMethodInfluential
Research works printed but not linked: 1
- D. Aysegul , J. Jonghoon , G. Vinayak , K. Bharadwaj , C. Alfredo , M. Berin , and C. Eugenio . Accelerating deep neural networks on mobile processor with embedded programmable logic . In NIPS 2013 . IEEE, 2013 . D. Aysegul, J. Jonghoon, G. Vinayak, K. Bharadwaj, C. Alfredo, M. Berin, and C. Eugenio. Accelerating deep neural networks on mobile processor with embedded programmable logic. In NIPS 2013. IEEE, 2013.
Cited by
Items 101–200 of 1,002
Efficient Processing of Deep Neural Networks
Sze, V., Chen, Y., Yang, T.-J., & Emer, J. (2020) · Synthesis lectures on computer architecture · doi:10.2200/s01004ed1v01y202004cac050 · Cited by 185
FPGA/DNN Co-Design: An Efficient Design Methodology for IoT Intelligence on the Edge
Hao, C., Zhang, X., Li, Y., Huang, S., Xiong, J., Rupnow, K., Hwu, W.-M., & Chen, D. (2019) · Proceedings of the 56th Annual Design Automation Conference 2019 · doi:10.1145/3316781.3317829 · Cited by 182
“The world has seen rapid adoption of FPGAs for DNN acceleration [1–6].”
A First Look at Deep Learning Apps on Smartphones
Xu, M., Liu, J., Liu, Y., Lin, F. X., Liu, Y., & Liu, X. (2019) · The World Wide Web Conference · doi:10.1145/3308558.3313591 · Cited by 182
“, model compression [50, 59, 67, 74, 97], hardware customizations [45, 47, 57, 103], and cloud offloading [61, 66, 99, 105].”
BackgroundBig Data Analytics for Large-scale Wireless Networks: Challenges and Opportunities
Dai, H.-N., Wong, R. C.-W., Wang, H., Zheng, Z., & Vasilakos, A. V. (2020) · ACM Computing Surveys · doi:10.1145/3337065 · Cited by 124
“For example, there are some efforts in FPGA-oriented Convolutional neural network (CNN) [233] and FPGA deep learning algorithm [234], which have the better performance than those implemented in common PCs.”
FPGA-based accelerator for long short-term memory recurrent neural networks
Guan, Y., Yuan, Z., Sun, G., & Cong, J. (2017) · 2017 22nd Asia and South Pacific Design Automation Conference (ASP-DAC) · doi:10.1109/aspdac.2017.7858394 · Cited by 180
“fixed-point computing units can achieve better performance and utilize less resource [19], the overall performance can even be better if we use fixed-point computing units instead of floating-point computing units.”
BackgroundSelf-citationMAGNet: A Modular Accelerator Generator for Neural Networks
Venkatesan, R., Shao, Y. S., Wang, M., Clemons, J., Dai, S., Fojtik, M., Keller, B., Klinefelter, A., Pinckney, N., Raina, P., et al. (2019) · 2019 IEEE/ACM International Conference on Computer-Aided Design (ICCAD) · doi:10.1109/iccad45719.2019.8942127 · Cited by 138
“Earlier research efforts have also proposed the use of FPGAs and corresponding automation tools to accelerate DNNs [14]–[16].”
MethodEfficient Hardware Architectures for Accelerating Deep Neural Networks: Survey
Pudi, D., Boppu, S., Manikandan, M. S., & Cenkeramaddi, L. R. (2022) · IEEE Access · doi:10.1109/access.2022.3229767 · Cited by 181
“No local reuse dataflow examples are found in [56] and [222].”
BackgroundMethodOptimizing CNN Model Inference on CPUsPreprint
Liu, Y., Wang, Y., Yu, R., Li, M., Sharma, V., & Wang, Y. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.02697 · Cited by 180
TANGRAM: Optimized Coarse-Grained Dataflow for Scalable NN Accelerators
Gao, M., Yang, X., Pu, J., Horowitz, M., & Kozyrakis, C. (2019) · Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems · doi:10.1145/3297858.3304014 · Cited by 178
Efficient and Effective Sparse LSTM on FPGA with Bank-Balanced Sparsity
Cao, S., Zhang, C., Yao, Z., Xiao, W., Nie, L., Zhan, D., Liu, Y., Wu, M., & Zhang, L. (2019) · Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays · doi:10.1145/3289602.3293898 · Cited by 178
“Hardware acceleration of DNNs has received significant attention from both industry and academia [4, 12, 16, 29].”
BackgroundSelf-citationAn Overview of Machine Learning within Embedded and Mobile Devices–Optimizations and Applications
Ajani, T. S., Imoize, A. L., & Atayero, A. A. (2021) · Sensors · doi:10.3390/s21134412 · Cited by 176
“Also, hardware design-based optimizations such as Tiling and data reuse may be utilized [25].”
MethodInfluentialSnaPEA: Predictive Early Activation for Reducing Computation in Deep Convolutional Neural Networks
Akhlaghi, V., Yazdanbakhsh, A., Samadi, K., Gupta, R. K., & Esmaeilzadeh, H. (2018) · 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca.2018.00061 · Cited by 173
Deep Neural Network Through an InP SOA-Based Photonic Integrated Cross-Connect
Shi, B., Calabretta, N., & Stabile, R. (2019) · IEEE Journal of Selected Topics in Quantum Electronics · doi:10.1109/jstqe.2019.2945548 · Cited by 176
“electronics: among those we name IBM TrueNorth [9], Neurogrid [10], SpiNNaker [11], and BrainDrop [12], for spiking neural networks, FPGA [13] and Google TPU [14] for deep neural networks.”
MethodElf: accelerate high-resolution mobile deep vision with content-aware parallel offloading
Zhang, W., He, Z., Liu, L., Jia, Z., Liu, Y., Gruteser, M., Raychaudhuri, D., & Zhang, Y. (2021) · Proceedings of the 27th Annual International Conference on Mobile Computing and Networking · doi:10.1145/3447993.3448628 · Cited by 172
“On the other hand, people have proposed to increase the computing resources by using massive accelerators, such as GPU, FPGA [14] and ASIC [15].”
MethodInfluentialViA: A Novel Vision-Transformer Accelerator Based on FPGA
Wang, T., Gong, L., Wang, C., Yang, Y., Gao, Y., Zhou, X., & Chen, H. (2022) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2022.3197489 · Cited by 145
“Because this one-size-fits-all approach may lead to the inconsistent final performance on network models with different workload characteristics [43], [44], this also appears on ViT, and there are many kinds of nonlinear functions between each computation part.”
BackgroundUsing Dataflow to Optimize Energy Efficiency of Deep Neural Network Accelerators
Chen, Y.-H., Emer, J., & Sze, V. (2017) · IEEE Micro · doi:10.1109/mm.2017.54 · Cited by 139
“This is to minimize DRAM access energy consumption by storing more data on-chip (for example, work by Tianshi Chen and colleagues 11 and Chen Zhang and colleagues 12 .”
MethodOPU: An FPGA-Based Overlay Processor for Convolutional Neural Networks
Yu, Y., Wu, C. H., Zhao, T., Wang, K., & He, L. (2019) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2019.2939726 · Cited by 172
“Roof-line model [5] reveals the relationship between bandwidth utilization and computational”
BackgroundMethodZheng, S., Liang, Y., Wang, S., Chen, R., & Sheng, K. (2020) · Proceedings of the Twenty-Fifth International Conference on Architectural Support for Programming Languages and Operating Systems · doi:10.1145/3373376.3378508 · Cited by 170
“On FPGA, the baseline uses schedules in [65] for performance optimization.”
MethodSharing, protection, and compatibility for reconfigurable fabric with Amorphos
Khawaja, A., Landgraf, J., Prakash, R., Wei, M., Schkufza, E., & Rossbach, C. J. (2018) · doi:10.5555/3291168.3291177 · Cited by 170
DOTA: detect and omit weak attentions for scalable transformer acceleration
Qu, Z., Liu, L., Tu, F., Chen, Z., Ding, Y., & Xie, Y. (2022) · Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems · doi:10.1145/3503222.3507738 · Cited by 130
“Finally, hardware accelerators for DNNs are related to executing the non-attention components of Transformers [6, 8ś10, 13ś15, 24ś 26, 30, 47, 55].”
MethodFathom: reference workloads for modern deep learning methods
Adolf, R., Rama, S., Reagen, B., Wei, G.-Y., & Brooks, D. (2016) · 2016 IEEE International Symposium on Workload Characterization (IISWC) · doi:10.1109/iiswc.2016.7581275 · Cited by 169
“Category Feature [8] [9] [10] [11] [12] [14] [21] [24] [26] [35] [38] [39] [40] [44] [47] [49] Fathom”
Origami: A 803-GOp/s/W Convolutional Network Accelerator
Cavigelli, L., & Benini, L. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2592330 · Cited by 169
“Catapult platform in [39] with very little known details and a HLS-based implementation [40] with a performance and energy efficiency inferior to nn-X.”
BackgroundUnderstanding the impact of precision quantization on the accuracy and energy of neural networks
Hashemi, S., Anthony, N., Tann, H., Bahar, R. I., & Reda, S. (2017) · Design, Automation, and Test in Europe · doi:10.23919/date.2017.7927224 · Cited by 170
ReBNet: Residual Binarized Neural Network
Ghasemzadeh, M., Samragh, M., & Koushanfar, F. (2018) · 2018 IEEE 26th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM) · doi:10.1109/fccm.2018.00018 · Cited by 132
“A line of research has therefore been focused on the development of FPGA accelerators for execution of CNN applications [4], [5], [6].”
MethodNon-Structured DNN Weight Pruning—Is It Beneficial in Any Platform?
Ma, X., Lin, S., Ye, S., He, Z., Zhang, L., Yuan, G., Tan, S. H., Li, Z., Fan, D., Qian, X., et al. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3063265 · Cited by 115
Efficient Hardware Architectures for Deep Convolutional Neural Network
Wang, J., Lin, J. T., & Wang, Z. (2017) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2017.2767204 · Cited by 166
ANT: Exploiting Adaptive Numerical Data Type for Low-bit Deep Neural Network Quantization
Guo, C., Zhang, C., Leng, J., Liu, Z., Yang, F., Liu, Y., Guo, M., & Zhu, Y. (2022) · IEEE/ACM International Symposium on Microarchitecture (MICRO) · doi:10.1109/micro56248.2022.00095 · Cited by 166
“For hardware acceleration, the proposed architectures are tailored to fit the computation characteristics of DNN models which leverage the regular access pattern, high data reuse and tremendous parallelism to save the area and latency from control logic [12], [14], [27], [28], [30], [36], [37], [51], [68], [69], [88],…”
MethodSelf-citationModeling Deep Learning Accelerator Enabled GPUs
Raihan, A., Goli, N., & Aamodt, T. M. (2019) · IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) · doi:10.1109/ispass.2019.00016 · Cited by 163
High-Performance FPGA-Based CNN Accelerator With Block-Floating-Point Arithmetic
Lian, X., Liu, Z., Song, Z., Dai, J., Zhou, W., & Ji, X. (2019) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2019.2913958 · Cited by 162
“[7], which can balance the computational throughput and memory bandwidth.”
BackgroundMethodCOMBA: A comprehensive model-based analysis framework for high level synthesis of real applications
Zhao, J., Feng, L., Sinha, S., Zhang, W., Liang, Y. E., & He, B. (2017) · International Conference on Computer Aided Design · doi:10.1109/iccad.2017.8203809 · Cited by 162
fpgaConvNet: Mapping regular and irregular convolutional neural networks on FPGAs
Venieris, S. I., & Bouganis, C.-S. (2019) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2018.2844093 · Cited by 161
Scalable and modularized RTL compilation of Convolutional Neural Networks onto FPGA
Ma, Y., Suda, N., Cao, Y., Seo, J.-S., & Vrudhula, S. (2016) · 2016 26th International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl.2016.7577356 · Cited by 161
“Previous FPGA implementations based on high-level synthesis tools have shown good scalability and require short design time, but their memory utilization is inefficient and may not allow exploitation of low-level hardware structures to achieve higher throughput [10][11][12].”
BackgroundMethodMoving Convolutional Neural Networks to Embedded Systems: The AlexNet and VGG-16 Case
Alippi, C., Disabato, S., & Roveri, M. (2018) · 2018 17th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN) · doi:10.1109/ipsn.2018.00049 · Cited by 161
“In this research stream, [46] introduced a FPGA-based accelerator for CNNs implementing floatingpoint operations for the convolutional layers.”
MethodDeep Learning on Mobile and Embedded Devices: State-of-the-art, Challenges, and Future Directions
Chen, Y., Zheng, B., Zhang, Z., Wang, Q., Shen, C., & Zhang, Q. (2021) · ACM Computing Surveys · doi:10.1145/3398209 · Cited by 159
“Zhang et al. [197] introduced a roofline-model-based method to optimize FPGA-based accelerator design.”
BackgroundMethodEvaluating Fast Algorithms for Convolutional Neural Networks on FPGAs
Liang, Y., Lu, L., Xiao, Q., & Yan, S. (2020) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2019.2897701 · Cited by 159
“Compared to prior work [13], we improve the average convolution performance from 61.6 to 1006.4 GOP/s. 1 For the overall CNN, we improve the performance from 72.4 to 854.6 GOP/s compared to [2].”
BackgroundMethodInfluentialFrequency Domain Acceleration of Convolutional Neural Networks on CPU-FPGA Shared Memory System
Zhang, C., & Prasanna, V. (2017) · Proceedings of the 2017 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays · doi:10.1145/3020078.3021727 · Cited by 158
“Table 7 shows that our design improves resource efficiency by 4.5x and 2.28x compared with state-of-the-art CNN implementations[21, 13].”
BackgroundMethodInfluentialSelf-citationUltracompact meta-imagers for arbitrary all-optical convolution
Fu, W., Zhao, D., Li, Z., Liu, S., Tian, C., & Huang, K. (2022) · Light: Science & Applications · doi:10.1038/s41377-022-00752-5 · Cited by 158
Device and materials requirements for neuromorphic computing
Islam, R., Li, H., Chen, P.-Y., Wan, W., Chen, H.-Y., Gao, B., Wu, H., Yu, S., Saraswat, K., & Philip Wong, H.-S. (2019) · Journal of Physics D: Applied Physics · doi:10.1088/1361-6463/aaf784 · Cited by 158
“4 caching, and processing reusable data in a DNN computation flow, and can be seen in several other reports of DNN accelerators as well [35, 36].”
MethodSC-DCNN: Highly-Scalable Deep Convolutional Neural Network using Stochastic Computing
Ren, A., Li, Z., Ding, C., Qiu, Q., Wang, Y., Li, J., Qian, X., & Yuan, B. (2017) · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3037697.3037746 · Cited by 155
“FPGA-based accelerators, benefited from the advantages of being programmable, high degree of parallelism and short develop round, is another promising path towards the hardware implementation of DCNNs [9, 10].”
BackgroundA Survey on Convolutional Neural Network Accelerators: GPU, FPGA and ASIC
Hu, Y., Liu, Y., & Liu, Z. (2022) · International Conference on Computer Research and Development · doi:10.1109/iccrd54409.2022.9730377 · Cited by 155
A New Stochastic Computing Multiplier with Application to Deep Convolutional Neural Networks
Sim, H., & Lee, J. (2017) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3061639.3062290 · Cited by 154
“Our SC-CNN accelerator architecture is by design very similar to previous CNN accelerators [15, 19] based on conventional binary.”
MethodMa, Y., Cao, Y., Vrudhula, S., & Seo, J.-S. (2017) · 2017 27th International Conference on Field Programmable Logic and Applications (FPL) · doi:10.23919/fpl.2017.8056824 · Cited by 134
“Therefore, the CNN acceleration strategy, such as loop unrolling and tiling techniques, must be optimized to properly manage the parallel computation and data storage patterns in memory [12][17][19].”
MethodAccuracy vs. Efficiency: Achieving Both through FPGA-Implementation Aware Neural Architecture Search
Jiang, W., Zhang, X., Sha, E. H.-M., Yang, L., Zhuge, Q., Shi, Y., & Hu, J. (2019) · Proceedings of the 56th Annual Design Automation Conference 2019 · doi:10.1145/3316781.3317757 · Cited by 121
“With such vision, a series of works for implementing neural networks on FPGAs have been carried out [3, 4, 8, 9, 11, 13, 15].”
BackgroundMethodInfluentialBrain-Inspired Computing: A Systematic Survey and Future Trends
Li, G., Deng, L., Tang, H., Pan, G., Tian, Y., Roy, K., & Maass, W. (2024) · Proceedings of the IEEE · doi:10.1109/jproc.2024.3429360 · Cited by 155
“Deep learning accelerators mainly refer to ANN accelerators [43], [171], [172], [173], [174], [175], [176], [177], [178], [179], [180], [181], [182], which use specific chips to accelerate the execution of ANNs.”
Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks
Gysel, P. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.06402 · Cited by 97
“ed accelerators have a shorter development time than ASICs, however they can’t match the throughput of GPUs. Dierent FPGA-based accelerators for neural networks have 21 been proposed. An approach by Zhang et al. (2015) uses Vivado HLS to accelerate the convolutional layers of AlexNet. Their oating point implementation…”
MethodSynetgy: Algorithm-hardware Co-design for ConvNet Accelerators on Embedded FPGAs
Yang, Y., Huang, Q., Wu, B., Zhang, T., Ma, L., Gambardella, G., Blott, M., Lavagno, L., Vissers, K., Wawrzynek, J., et al. (2019) · Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays · doi:10.1145/3289602.3293902 · Cited by 102
“[28] quantitatively studies the computation throughput and memory bandwidth requirement for ConvNets.”
BackgroundMix and Match: A Novel FPGA-Centric Deep Neural Network Quantization Framework
Chang, S.-E., Li, Y., Sun, M., Shi, R., So, H. K.-H., Qian, X., Wang, Y., & Lin, X. (2021) · 2021 IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca51647.2021.00027 · Cited by 126
“Extensive research efforts have been dedicated to the efficient implementations of DNN inference models on various edgecomputing platforms, such as ASICs [8]–[14], FPGAs [15]– [18], and embedded CPUs/GPUs [19]–[23].”
BackgroundAn Efficient Hardware Accelerator for Structured Sparse Convolutional Neural Networks on FPGAs
Zhu, C., Huang, K., Yang, S., Zhu, Z., Zhang, H., & Shen, H. (2020) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2020.3002779 · Cited by 150
“In this section, we adopt the well-known roofline model [15] for exploring the impact of insufficient off-chip bandwidth on performance.”
BackgroundMethodConfuciuX: Autonomous Hardware Resource Assignment for DNN Accelerators using Reinforcement Learning
Kao, S.-C., Jeong, G., & Krishna, T. (2020) · IEEE/ACM International Symposium on Microarchitecture (MICRO) · doi:10.1109/micro50266.2020.00058 · Cited by 150
“For this model, we assume an underlying accelerator can heterogeneously partition the (PE, buffer) resources at either design-time (e.g., ASIC [9], [90], [96]) or compile-time (e.g., CGRA [39]/FPGA [92]).”
BackgroundInfluentialHnamte, V., & Hussain, J. (2023) · Telematics and Informatics Reports · doi:10.1016/j.teler.2023.100077 · Cited by 149
Neuromorphic Computing Based on Emerging Memory Technologies
Rajendran, B., & Alibart, F. (2016) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2016.2533298 · Cited by 148
“Recently, there have been proposals to accelerate these algorithms in FPGA technology [49].”
BackgroundA Multi-Neural Network Acceleration Architecture
Baek, E., Kwon, D., & Kim, J. (2020) · 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca45697.2020.00081 · Cited by 107
“For example, weight-stationary [7], [13], [16], [33], [43], [46], [47], [51], [52], [59], output-stationary [22], [51], row-stationary [14], [15], [26] dataflows determine spatial and temporal data mappings to PEs differently.”
F-CNN: An FPGA-based framework for training Convolutional Neural Networks
Zhao, W., Fu, H., Luk, W. W., Yu, T., Wang, S., Feng, B., Ma, Y., & Yang, G. (2016) · IEEE International Conference on Application-specific Systems, Architectures and Processors (ASAP) · doi:10.1109/asap.2016.7760779 · Cited by 147
“Ovtcharov [11] demonstrated a 3x better performance on convolutional accelerator than [4], but did not provide more details on the design.”
BackgroundMethodInfluentialHigh-Performance CNN Accelerator on FPGA Using Unified Winograd-GEMM Architecture
Kala, S., Jose, B. R., Mathew, J., & Nalesh, S. (2019) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2019.2941250 · Cited by 147
“Memory bandwidth issues in the CNNs are discussed in [18].”
BackgroundPerformance Modeling and Evaluation of Distributed Deep Learning Frameworks on GPUs
Shi, S., Wang, Q., & Chu, X. (2018) · IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech) · doi:10.1109/dasc/picom/datacom/cyberscitec.2018.000-4 · Cited by 146
Cloud-DNN: An Open Framework for Mapping DNN Models to Cloud FPGAs
Chen, Y., He, J., Zhang, X., Hao, C., & Chen, D. (2019) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3289602.3293915 · Cited by 143
“Based on our experimental results and previous literature [20, 21], the LUT and FF are not the bottleneck for accelerator system generation.”
InfluentialLaconic deep learning inference acceleration
Sharify, S., Lascorz, A. D., Mahmoud, M., Nikolić, M., Siu, K., Stuart, D. M., Poulos, Z., & Moshovos, A. (2019) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3307650.3322255 · Cited by 142
“The first class includes dataflow scheduling and data blocking which exploit data reuse to reduce the energy needed by data movement [14, 30, 52, 55, 65, 92, 95, 96, 100].”
BackgroundFLAT: An Optimized Dataflow for Mitigating Attention Bottlenecks
Kao, S.-C., Subramanian, S., Agrawal, G., Yazdanbakhsh, A., & Krishna, T. (2023) · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3575693.3575747 · Cited by 145
“This is because the main operators within attention layers exhibit distinct compute and memory characteristics posing notable bottlenecks on off-chip memory bandwidth compared to CONV and FC.”
BackgroundInfluentialAn IoT Endpoint System-on-Chip for Secure and Energy-Efficient Near-Sensor Analytics
Conti, F., Schilling, R., Schiavone, P. D., Pullini, A., Rossi, D., Gürkaynak, F. K., Muehlberghuber, M., Gautschi, M., Loi, I., Haugou, G., et al. (2017) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2017.2698019 · Cited by 147
“The most common way to accelerate CNNs is to rely on powerful GP-GPUs [7], [19] or on FPGAs [20], [21].”
MethodCentaur: A Chiplet-based, Hybrid Sparse-Dense Accelerator for Personalized Recommendations
Hwang, R., Kim, T., Kwon, Y., & Rhu, M. (2020) · 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca45697.2020.00083 · Cited by 99
Dave, S., Baghdadi, R., Nowatzki, T., Avancha, S., Shrivastava, A., & Li, B. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3098483 · Cited by 111
“models [35], [36], [80], [143].”
MethodHyPar: Towards Hybrid Parallelism for Deep Learning Accelerator Array
Song, L., Mao, J., Zhuo, Y., Qian, X., Li, H., & Chen, Y. (2019) · 2019 IEEE International Symposium on High Performance Computer Architecture (HPCA) · doi:10.1109/hpca.2019.00027 · Cited by 108
DeepWear: Adaptive Local Offloading for On-Wearable Deep Learning
Xu, M., Qian, F., Zhu, M., Huang, F., Pushp, S., & Liu, X. (2020) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2019.2893250 · Cited by 138
“Some other efforts such as [31], [32], [33], [34] aimed at building customized hardware accelerators for DL or other machine learning tasks.”
BackgroundProgramming Heterogeneous Systems from an Image Processing DSL
Pu, J., Bell, S., Yang, X., Setter, J., Richardson, S., Ragan‐Kelley, J., & Horowitz, M. (2017) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3107953 · Cited by 140
“Configurable hardware, using either coarse grain reconfigurable arrays (CGRA) [15, 9] or FPGAs [12] are one approach to providing a flexible machine that can be config-ured to match the data flow of different algorithms.”
MethodInfluentialMiyashita, D., Kousai, S., Suzuki, T., & Deguchi, J. (2017) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2017.2712626 · Cited by 137
“For example, optimizing ordering of computation so that the number of memory accesses is minimized allows us to reduce the overall energy consumption significantly [21]–[24].”
BackgroundThroughput-Optimized FPGA Accelerator for Deep Convolutional Neural Networks
Liu, Z., Dou, Y., Jiang, J., Xu, J., Li, S., Zhou, Y., & Xu, Y. (2017) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3079758 · Cited by 116
“Zhang et al. [2015] implements an HLS-based FPGA accelerator.”
BackgroundResultInfluentialChang, J.-W., Kang, K.-W., & Kang, S. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2888898 · Cited by 136
In-Situ AI: Towards Autonomous and Incremental Deep Learning for IoT Systems
Song, M., Zhong, K., Zhang, J., Hu, Y., Liu, D., Zhang, W., Wang, J., & Li, T. (2018) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca.2018.00018 · Cited by 131
“[29] propose an analytical design scheme based on the roofline model to find the fastest FPGA implementation for CONV layers.”
MethodGong, L., Wang, C., Li, X., Chen, H., & Zhou, X. (2018) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2018.2857078 · Cited by 118
“Here, the off-chip memory bandwidth of mainstream FPGA devices we measured by the method proposed in [8].”
MethodSun, M., Li, Z., Lu, A., Li, Y., Chang, S.-E., Ma, X., Lin, X., & Fang, Z. (2022) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3490422.3502364 · Cited by 133
“Our base FPGA accelerator design is similar to [50] shortcut connection).”
MethodUnderstanding the Potential of FPGA-based Spatial Acceleration for Large Language Model Inference
Chen, H., Zhang, J., Du, Y., Xiang, S., Yue, Z., Zhang, N., Cai, Y., & Zhang, Z. (2024) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3656177 · Cited by 136
“These two stages exhibit significantly different computational and memory characteristics (detailed in Section 3), making it necessary to tailor hardware accelerators for their specific needs.”
Big/little deep neural network for ultra low power inference
Park, E., Kim, D., Kim, S., Kim, Y.-D., Kim, G., Yoon, S., & Yoo, S. (2015) · 2015 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS) · doi:10.1109/codesisss.2015.7331375 · Cited by 130
“We adopted a tile-based architecture in [10, 32] and ran maximum 512 neurons in parallel.”
BackgroundMethodInfluentialOptimizing Memory Efficiency for Deep Convolutional Neural Networks on GPUs
Li, C., Yang, Y., Feng, M., Chakradhar, S., & Zhou, H. (2016) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1109/sc.2016.53 · Cited by 130
DRQ: Dynamic Region-based Quantization for Deep Neural Network Acceleration
Song, Z., Fu, B., Wu, F., Jiang, Z., Jiang, L., Jing, N., & Liang, X. (2020) · 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca45697.2020.00086 · Cited by 91
“In the last few years, various NN accelerators are proposed for inference [4], [6], [8], [11], [13], [18], [28], [30], [36].”
BackgroundNeurostream: Scalable and Energy Efficient Deep Learning with Smart Memory Cubes
Azarkhish, E., Rossi, D., Loi, I., & Benini, L. (2017) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2017.2752706 · Cited by 129
“We can observe that even on the most powerful GPUs, the utilization of computing resources does not exceed 55%, which is directly reflected in a lower energy-efficiency.”
BackgroundMethodInfluentialLight-OPU: An FPGA-based Overlay Processor for Lightweight Convolutional Neural Networks
Yu, Y., Zhao, T., Wang, K., & He, L. (2020) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3373087.3375311 · Cited by 130
“Conventional convolutional neural network (CNN) acceleration on FPGA has drawn much attention in recent years [2, 4, 7, 16, 18, 21, 24, 28, 34, 37–39].”
BackgroundEmbedded Streaming Deep Neural Networks Accelerator With Applications
Dundar, A., Jin, J., Martini, B., & Culurciello, E. (2017) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2016.2545298 · Cited by 120
“analytical design scheme using roofline model [36].”
Zhao, T., Xie, Y., Wang, Y., Cheng, J., Guo, X., Hu, B., & Chen, Y. (2022) · Proceedings of the IEEE · doi:10.1109/jproc.2022.3153408 · Cited by 92
REQ-YOLO: A Resource-Aware, Efficient Quantization Framework for Object Detection on FPGAs
Ding, C., Wang, S., Liu, N., Xu, K., Wang, Y., & Liang, Y. E. (2019) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3289602.3293904 · Cited by 126
“The second one is efficient hardware implementations, including FPGAs and ASICs [1, 7, 31, 32, 36, 49, 50, 53, 54].”
MethodLUTNet: Rethinking Inference in FPGA Soft Logic
Wang, E., Davis, J. J., Cheung, P. Y. K., & Constantinides, G. A. (2019) · IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM) · doi:10.1109/fccm.2019.00014 · Cited by 128
Maximizing CNN Accelerator Efficiency Through Resource Partitioning
Shen, Y., Ferdman, M., & Milder, P. A. (2017) · ACM SIGARCH Computer Architecture News · doi:10.1145/3140659.3080221 · Cited by 124
DNNExplorer: a framework for modeling and exploring a novel paradigm of FPGA-based DNN accelerator
Zhang, X., Ye, H., Wang, J., Lin, Y., Xiong, J., Hwu, W., & Chen, D. (2020) · doi:10.1145/3400302.3415609 · Cited by 124
“There are rich studies on customized DNN accelerators that take advantage of different hardware devices, such as exploring kernel optimizations on GPUs and customizing accelerators on ASICs and FPGAs [12–16].”
BackgroundCaffeinated FPGAs: FPGA framework For Convolutional Neural Networks
DiCecco, R., Lacey, G., Vasiljevic, J., Chow, P., Taylor, G., & Areibi, S. (2016) · 2016 International Conference on Field-Programmable Technology (FPT) · doi:10.1109/fpt.2016.7929549 · Cited by 103
“Metric Zhang [1] Suda [2] Qiu [3] This Work Clock Frequency (MHz) 100 120 150 200 Precision 32 bit float 8-16-bit fixed 16-bit fixed 32 bit float FPGA Version Virtex 7 VX485T Stratix-V GSD8 Zynq XC7Z045 Virtex 7 XC7VX690T-2 DSP Utilization 2,240 (Not specified) 780 1,307 Host Connection Microblaze, on chip host PCIe A…”
BackgroundSong, L., Wang, Y., Han, Y., Zhao, X., Liu, B., & Li, X. (2016) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/2897937.2897995 · Cited by 122
“2 is a typical architecture of state-of-the-art deep learning accelerators [8, 13, 14], which consists of four main components: one input data buffer, one output data buffers, one weight buffer, a computational block (neural Processing Unit, PE) and a logic Control Unit (CU).”
BackgroundMethodResultInfluentialLatency-driven design for FPGA-based convolutional neural networks
Venieris, S. I., & Bouganis, C.-S. (2017) · 2017 27th International Conference on Field Programmable Logic and Applications (FPL) · doi:10.23919/fpl.2017.8056828 · Cited by 91
“With respect to AlexNet, we compare with the roofline model-based design by [11] and the automated design flows of [12] and [13] which are the closest works to our automated approach.”
MethodAll-optical graph representation learning using integrated diffractive photonic computing units
Yan, T., Yang, R., Zheng, Z., Lin, X., Xiong, H., & Dai, Q. (2022) · Science Advances · doi:10.1126/sciadv.abn7630 · Cited by 122
“The core is to leverage multi-layer neural networks to learn hierarchical and complicated abstracts from big data, driven by the continuous development of integrated electronic computing platforms, such as CPUs 7 , GPUs 8 , TPUs 9 , and FPGAs 10 .”
BackgroundSextans: A Streaming Accelerator for General-Purpose Sparse-Matrix Dense-Matrix Multiplication
Song, L., Chi, Y., Sohrabizadeh, A., Choi, Y., Lau, J., & Cong, J. (2022) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3490422.3502357 · Cited by 121
“SpMM, a memory-intensive application which is distinguished from typical computation-intensive FPGA applications [5, 19, 27, 68, 78, 79], is a good fit for HBM.”
Self-citationEfficient FPGA Acceleration of Convolutional Neural Networks Using Logical-3D Compute Array
Rahman, A., Lee, J., & Choi, K. (2016) · Design, Automation, and Test in Europe · doi:10.3850/9783981537079_0833 · Cited by 120
“#DSPs/MAC 5 4 3 2 1 DSP utilization of [8] 91.”
BackgroundMethodResultInfluentialRaPiD: AI Accelerator for Ultra-low Precision Training and Inference
Venkataramani, S., Srinivasan, V., Wang, W., Sen, S., Zhang, J., Agrawal, A., Kar, M., Jain, S., Mannari, A., Tran, H., et al. (2021) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1109/isca52012.2021.00021 · Cited by 122
“Recognizing this, a myriad of accelerators ranging from low-power ASIC / FGPA / CGRA cores [21, 22, 23, 24, 26, 32, 47, 71, 72, 73, 74, 75, 76, 77, 78] to large scale systems [29, 30, 31, 79, 80, 81] have been proposed.”
BackgroundA Survey of Design and Optimization for Systolic Array-based DNN Accelerators
Xu, R., Ma, S., Guo, Y., & Li, D. (2023) · ACM Computing Surveys · doi:10.1145/3604802 · Cited by 120
HolyLight: A Nanophotonic Accelerator for Deep Learning in Data Centers
Liu, W., Liu, W., Ye, Y., Lou, Q., Xie, Y., & Jiang, L. (2019) · Design, Automation, and Test in Europe · doi:10.23919/date.2019.8715195 · Cited by 119
“We selected an Intel Xeon CPU, an Nvidia Tesla GPU, a Xilinx Virtex7 FPGA [29], a DaDianNao ASIC chip [4], a Google TPU [9], and a ReRAM-based CNN accelerator ISAAC [19].”
BackgroundMethodCharacterizing the deployment of deep neural networks on commercial edge devices
Hadidi, R., Cao, J., Xie, Y., Asgari, B., Krishna, T., & Kim, H. (2019) · 2019 IEEE International Symposium on Workload Characterization (IISWC) · doi:10.1109/iiswc47752.2019.9041955 · Cited by 118
Hardware accelerators for recurrent neural networks on FPGA
Chang, A. X. M., & Culurciello, E. (2017) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas.2017.8050816 · Cited by 118
Design Flow of Accelerating Hybrid Extremely Low Bit-Width Neural Network in Embedded FPGA
Wang, J., Lou, Q., Zhang, X., Zhu, C., Lin, Y., & Chen, D. (2018) · 2018 28th International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl.2018.00035 · Cited by 104
An FPGA-Based CNN Accelerator Integrating Depthwise Separable Convolution
Liu, B., Zou, D., Feng, L., Feng, S., Fu, P., & Li, J. (2019) · Electronics · doi:10.3390/electronics8030281 · Cited by 117
EdgeLLM: A Highly Efficient CPU-FPGA Heterogeneous Edge Accelerator for Large Language Models
Huang, M., Shen, A., Li, K., Peng, H., Li, B., Su, Y., & Yu, H. (2025) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2025.3546256 · Cited by 127
“Figure 3 shows the roofline model [35] when accelerating the matrix multiplication (MatMUL) operator, serving as a pivotal analytical tool to contrast the boundaries of memory-limited and computation-limited domains.”
EVA²: Exploiting Temporal Redundancy in Live Computer Vision
Buckler, M., Bedoukian, P., Jayasuriya, S., & Sampson, A. (2018) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1109/isca.2018.00051 · Cited by 115
“In research, many recent ASICs target convolutional and fullyconnected layers [43, 17, 1, 44, 2, 45, 46, 39], as do some FPGA designs [47, 48, 49, 50].”
BackgroundGroundwork for Neural Network-Based Specific Emitter Identification Authentication for IoT
McGinthy, J. M., Wong, L. J., & Michaels, A. J. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2908759 · Cited by 116
“Efficient hardware implementations of NNs is also an active area of research that could also be applied to both reduce latency and power consumption [77].”
BackgroundSmartShuttle: Optimizing off-chip memory accesses for deep learning accelerators
Li, J., Yan, G., Lü, W., Jiang, S., Gong, S., Wu, J., & Li, X. (2018) · Design, Automation, and Test in Europe · doi:10.23919/date.2018.8342033 · Cited by 115
“To make a fair comparison, our implementation has two versions, one with data compression (SmartShuttle, to compare with Eyeriss) and another without data compression (SmartShuttle-CF, compression free, to compare with Zhang).”
BackgroundMethodInfluentialA Framework for Generating High Throughput CNN Implementations on FPGAs
Zeng, H., Chen, R., Zhang, C., & Prasanna, V. K. (2018) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3174243.3174265 · Cited by 114
“Work in [4] proposed a roofline model to capture various techniques including loop tiling, unrolling and interchanging.”
BackgroundMethodInfluentialSelf-citation
Metadata
- Authors
- Zhang, C., Li, P., Sun, G., Guan, Y., Xiao, B., & Cong, J.
- Affiliations
- Peking University (CN)
- University of California, Los Angeles (US)
- Venue
- Proceedings of the 2015 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays
- Publisher
- ACM
- Published
- 2015-02-22
- Pages
- 161-170
- OpenAlex
- W2094756095
- Cited by
- 2,158
- References
- 17
- Citations by year
- 2026: 78 · 2025: 116 · 2024: 145 · 2023: 162 · 2022: 233 · 2021: 253 · 2020: 258 · 2019: 303 · 2018: 281 · 2017: 221 · 2016: 90 · 2015: 15
- FWCI
- 133.92
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) (publishedVersion)
- Funders
- C-FAR
- NSF China (61202072)
- National High Technology Research and Development Program of China (2012AA010902)
- RFDP (20110001110099)
- Topics
- Physical Sciences › Computer Science › Hardware and Architecture › Embedded Systems Design Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Embedded Systems and FPGA Applications
- Keywords
- FPGA accelerator, deep convolutional neural networks, roofline model, memory bandwidth, loop tiling, loop transformation, FPGA resource utilization, image 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.