Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 496

The DOI matches its Crossref record; 29 of the 29 references it cites were found in a registry or an index.

FPGA-Based Accelerators of Deep Learning Networks for Learning and Classification: A Review

Open access
Shawahna, Ahmad, Sait, Sadiq M., El-Maleh, Aiman·2019·IEEE Access
DOI 10.1109/access.2018.2890150
Open this paper in CoChat — free account

Abstract

Due to recent advances in digital technologies, and availability of credible data, an area of artificial intelligence, deep learning, has emerged and has demonstrated its ability and effectiveness in solving complex learning problems not possible before. In particular, convolutional neural networks (CNNs) have demonstrated their effectiveness in the image detection and recognition applications. However, they require intensive CPU operations and memory bandwidth that make general CPUs fail to achieve the desired performance levels. Consequently, hardware accelerators that use application-specific integrated circuits, field-programmable gate arrays (FPGAs), and graphic processing units have been employed to improve the throughput of CNNs. More precisely, FPGAs have been recently adopted for accelerating the implementation of deep learning networks due to their ability to maximize parallelism and their energy efficiency. In this paper, we review the recent existing techniques for accelerating deep learning networks on FPGAs. We highlight the key features employed by the various techniques for improving the acceleration performance. In addition, we provide recommendations for enhancing the utilization of FPGAs for CNNs acceleration. The techniques investigated in this paper represent the recent trends in the FPGA-based accelerators of deep learning networks. Thus, this paper is expected to direct the future advances on efficient hardware accelerators and to be useful for deep learning researchers.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/access.2018.2890150

Cites

29 of 29 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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

Show all 29 references

Research works printed but not linked: 183

  1. 10.1145/3079758 · DOI
  2. abadi, 2016, TensorFlow: A system for large-scale machine learning, Proc OSDI, 16, 265
  3. 10.1109/FCCM.2017.25 · DOI · Paper page
  4. chetlur, 2014, cuDNN Efficient Primitives for Deep Learning
  5. alsuwaiyel, 2016, Algorithms Design Techniques and Analysis, 14
  6. zaremba, 2014, Recurrent Neural Network Regularization
  7. kim, 2016, Bitwise Neural Networks
  8. rastegari, 2016, XNOR-Net: ImageNet classification using binary convolutional neural networks, Proc Eur Conf Comput Vis, 525
  9. 10.1016/j.vlsi.2017.12.009 · DOI
  10. lin, 2013, Network in Network
  11. mcnelis, 2005, Neural networks in Finance Gaining predictive edge in the market
  12. blanco-filgueira, 2018, Deep learning-based multiple object visual tracking on embedded system for IoT and mobile edge computing applications
  13. 10.1109/VLSIC.2005.1469371 · DOI
  14. 10.1109/CVPR.2015.7298594 · DOI · Paper page
  15. sermanet, 2009, A multirange architecture for collision-free off-road robot navigation, J Field Robot, 26, 52, 10.1002/rob.20270 · DOI
  16. hadsell, 2007, A multi-range vision strategy for autonomous offroad navigation, Proc Robot Appl (RA), 1, 457
  17. muller, 2006, Off-road obstacle avoidance through end-to-end learning, Proc Adv Neural Inf Process Syst, 739
  18. 10.1109/CVPR.2015.7299170 · DOI · Paper page
  19. 10.1145/3020078.3021744 · DOI · Paper page
  20. courbariaux, 2016, Binarized neural networks Training deep neural networks with weights and activations constrained to +1 or ?1
  21. 10.1137/1.9781611970364 · DOI · Paper page
  22. 10.1109/CVPR.2016.435 · DOI · Paper page
  23. 10.23919/FPL.2017.8056828 · DOI
  24. 10.1109/FCCM.2017.64 · DOI · Paper page
  25. 10.1145/3020078.3021738 · DOI · Paper page
  26. 10.1145/3020078.3021727 · DOI · Paper page
  27. 10.1137/1.9781611970999 · DOI · Paper page
  28. 10.1109/CVPR.2006.200 · DOI · Paper page
  29. zhou, 2016, DoReFa-Net Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
  30. simonyan, 2014, Very Deep Convolutional Networks for Large-scale Image Recognition
  31. 10.1109/TASLP.2014.2339736 · DOI · Paper page
  32. lai, 2015, Recurrent convolutional neural networks for text classification, Proc AAAI, 333, 2267
  33. 10.1109/ICDAR.2003.1227801 · DOI · Paper page
  34. 10.3115/v1/D14-1181 · DOI · Paper page
  35. 10.1109/TASLP.2014.2303296 · DOI · Paper page
  36. 2018, Imagenet Large Scale Visual Recognition Challenge
  37. 10.1109/TASL.2011.2109382 · DOI · Paper page
  38. 10.1109/92.784091 · DOI
  39. 2011, The OpenCL Specification Version 1 1
  40. 10.1145/2897937.2898003 · DOI · Paper page
  41. ovtcharov, 2015, Accelerating deep convolutional neural networks using specialized hardware, 2, 1
  42. 10.1145/2435264.2435273 · DOI · Paper page
  43. 10.1109/ICCD.2013.6657019 · DOI · Paper page
  44. 10.1109/TCSVT.2008.918775 · doi:10.1109/tcsvt.2008.918775
  45. 10.1109/CVPRW.2014.106 · DOI · Paper page
  46. 10.1109/ICCVW.2009.5457611 · DOI
  47. 10.1109/ASAP.2017.7995254 · DOI
  48. 10.1145/2647868.2654889 · DOI · Paper page
  49. 10.1145/2830772.2830810 · doi:10.1145/2830772.2830810
  50. 10.1145/2684746.2689060 · DOI · Paper page
  51. chilimbi, 2014, Project Adam: Building an efficient and scalable deep learning training system, Proc OSDI, 14, 571
  52. nomura, 2007, Projection-field-type VLSI convolutional neural networks using merged/mixed analog-digital approach, Proc Int Conf Neural Inf Process, 1081
  53. 10.1007/978-1-4471-0487-2 · DOI
  54. 10.1145/2744769.2744788 · DOI
  55. reeves, 1995, Modern Heuristic Techniques for Combinatorial Problems
  56. 10.1109/FCCM.2016.22 · DOI · Paper page
  57. 10.1109/PROC.1987.13876 · DOI · Paper page
  58. 10.1145/2435264.2435271 · DOI
  59. 10.1145/2966986.2967011 · DOI · Paper page
  60. whaley, 1998, Automatically tuned linear algebra software, Proc 1998 ACM/IEEE Supercomputing 98 Conf, 1
  61. 2018, DE5-Net FPGA Kit User Manual
  62. 10.1002/widm.1253 · DOI · Paper page
  63. goodfellow, 2016, Deep Learning, 1
  64. rumelhart, 1988, Learning representations by back-propagating errors, Neurocomputing Foundations of Research, 696, 10.7551/mitpress/4943.003.0042 · DOI
  65. weyand, 2016, PlaNet—Photo geolocation with convolutional neural networks, Proc Eur Conf Comput Vis, 37
  66. 2018, P395-D8 OpenCL FPGA Accelerator Cards
  67. nielsen, 2015, Neural Networks and Deep Learning, 25
  68. 10.1145/2664666.2664670 · doi:10.1145/2664666.2664670
  69. 2018, What is Deep Learning?
  70. 2018, OpenCL Design Examples
  71. 10.1109/CVPRW.2017.60 · DOI · Paper page
  72. zeiler, 2014, Visualizing and understanding convolutional networks, Proc Eur Conf Comput Vis, 818
  73. 10.1109/ICASSP.2013.6639346 · DOI · Paper page
  74. 10.1109/MLSP.2008.4685487 · DOI · Paper page
  75. 10.1002/rob.20276 · DOI · Paper page
  76. 10.1109/MM.2011.89 · DOI · Paper page
  77. 10.1145/1816038.1815968 · DOI
  78. 10.1145/1950413.1950423 · DOI · Paper page
  79. joshi, 2018, What is Local Response Normalization in Convolutional Neural Networks
  80. cong, 2014, Minimizing computation in convolutional neural networks, Proc Int Conf Artif Neural Netw, 281
  81. 10.1109/TPAMI.2007.56 · DOI · Paper page
  82. ma, 2016, Scalable and modularized RTL compilation of convolutional neural networks onto FPGA, Proc 16th Int Conf Field Program Logic Appl (FPL), 1
  83. 10.1145/197405.197406 · DOI · Paper page
  84. 10.1109/TCAD.2017.2705069 · DOI · Paper page
  85. 10.1016/j.neucom.2010.03.021 · DOI · Paper page
  86. 10.1145/3020078.3021740 · DOI · Paper page
  87. 10.1109/MICRO.2012.48 · DOI · Paper page
  88. 10.1145/3007787.3001163 · DOI · Paper page
  89. 10.1109/TCSI.2017.2735490 · DOI · Paper page
  90. 10.1007/978-1-4614-1791-0 · doi:10.1007/978-1-4614-1791-0
  91. 10.1145/2678373.2665678 · DOI · Paper page
  92. 10.1155/2012/649057 · doi:10.1155/2012/649057
  93. 10.1109/TCAD.2011.2110592 · DOI · Paper page
  94. wang, 2017, DLAU: A scalable deep learning accelerator unit on FPGA, IEEE Trans Comput -Aided Design Integr Circuits Syst, 36, 513
  95. 2018, JTAG UART Core
  96. 10.23919/FPL.2017.8056824 · DOI
  97. iandola, 2016, SqueezeNet AlexNet-level accuracy with 50 $\times$ fewer parameters and
  98. li, 2016, A high performance FPGA-based accelerator for large-scale convolutional neural networks, Proc 16th Int Conf Field Program Logic Appl (FPL), 1
  99. 10.1109/ISCAS.2017.8050344 · DOI
  100. 10.1007/0-387-28487-7 · DOI · Paper page
  101. 10.1109/MCSE.2010.69 · DOI · Paper page
  102. 10.1145/3020078.3021698 · DOI · Paper page
  103. 10.1109/HOTCHIPS.2009.7478342 · DOI
  104. czajkowski, 2012, OpenCL for FPGAs: Prototyping a compiler, Proc Int Conf Eng Reconfigurable Syst Algorithms (ERSA), 1
  105. 10.1017/CBO9781139042918.020 · DOI
  106. 10.1145/3079856.3080221 · DOI · Paper page
  107. lacey, 2016, Deep learning on FPGAs Past present and future
  108. 10.1007/978-981-10-0591-6 · doi:10.1007/978-981-10-0591-6
  109. 10.1145/2847263.2847265 · DOI · Paper page
  110. 10.1145/3020078.3021745 · DOI · Paper page
  111. 10.1007/978-3-319-68161-0 · doi:10.1007/978-3-319-68161-0
  112. 10.1109/CICC.2017.7993626 · DOI · Paper page
  113. nair, 2010, Rectified linear units improve restricted boltzmann machines, Proc 27th Int Conf Mach Learn (ICML), 807
  114. denil, 2013, Predicting parameters in deep learning, Proc Adv Neural Inf Process Syst, 2148
  115. karpathy, 2018, Convolutional Neural Network for Visual Recognition
  116. 10.1145/3020078.3021736 · DOI · Paper page
  117. 10.1145/2847263.2847276 · DOI · Paper page
  118. 10.1109/MCSE.2008.143 · DOI
  119. 10.1038/scientificamerican0697-66 · DOI
  120. brown, 2012, Field-Programmable Gate Arrays, 180
  121. abdelfattah, 2018, DLA Compiler and FPGA overlay for neural network inference acceleration
  122. van loan, 1996, Matrix Computations
  123. chung, 2017, Accelerating persistent neural networks at datacenter scale, Proc Hot Chips, 27
  124. 10.1109/TVLSI.2018.2815603 · DOI · Paper page
  125. 10.1109/FCCM.2015.15 · doi:10.1109/fccm.2015.15
  126. liu, 2016, SSD: Single shot MultiBox detector, Proc Eur Conf Comput Vis, 21
  127. 10.1109/ICCV.2017.154 · DOI · Paper page
  128. 10.1155/2016/1537325 · DOI
  129. 10.1145/2934583.2934644 · DOI · Paper page
  130. sait, 1999, Iterative Computer Algorithms with Applications in Engineering Solving Combinatorial Optimization Problems
  131. 10.1109/FPL.2009.5272559 · DOI · Paper page
  132. 10.1109/TNN.2006.883002 · DOI
  133. 10.1109/TCSII.2006.886898 · DOI
  134. 10.1007/s11265-005-4961-3 · DOI
  135. 10.1109/ASAP.2009.25 · DOI · Paper page
  136. lecun, 2015, Lenet-5 convolutional neural networks
  137. graf, 2009, A massively parallel digital learning processor, Proc Adv Neural Inf Process Syst, 529
  138. platt, 1999, 12 fast training of support vector machines using sequential minimal optimization, Advances in Kernel Methods, 185
  139. 10.1007/s10791-009-9117-9 · DOI
  140. 10.1109/FCCM.2009.34 · DOI
  141. 10.1145/1854273.1854309 · DOI
  142. sato, 1996, Generalized learning vector quantization, Proc Adv Neural Inf Process Syst, 423
  143. macqueen, 1967, Some methods for classification and analysis of multivariate observations, Proc 5th Berkeley Symp Math Statist Probab, 1, 281
  144. chellapilla, 2006, High performance convolutional neural networks for document processing, Proc Int'l Workshop Frontiers in Handwriting Recognition
  145. 10.1109/72.554195 · DOI · Paper page
  146. nasse, 2009, Face detection using GPU-based convolutional neural networks, Proc Int Conf Comput Anal Images Patterns, 83
  147. 10.1090/S0025-5718-1981-0595059-1 · DOI
  148. montgomery, 1994, A survey of modern integer factorization algorithms, CWI Quart, 7, 337
  149. 10.1109/ISCAS.2010.5537908 · DOI · Paper page
  150. 10.1109/CVPRW.2011.5981829 · DOI · Paper page
  151. 10.1016/j.neunet.2014.09.003 · DOI · Paper page
  152. collobert, 2008, Torch, Proc Workshop Mach Learn Open Source Softw (NIPS), 76, 113
  153. 10.1561/2200000006 · DOI · Paper page
  154. grangier, 2009, Deep convolutional networks for scene parsing, Proc ICML Deep Learn Workshop, 3, 109
  155. chen, 2014, DianNao: A small-footprint high-throughput accelerator for ubiquitous machine-learning, ACM SIGPLAN Notices, 49, 269, 10.1145/2644865.2541967 · DOI · Paper page
  156. han, 2015, Deep compression Compressing deep neural networks with pruning trained quantization and huffman coding
  157. hassibi, 1993, Second order derivatives for network pruning: Optimal brain surgeon, Proc Adv Neural Inf Process Syst, 164
  158. hanson, 1989, Comparing biases for minimal network construction with back-propagation, Proc Adv Neural Inf Process Syst, 177
  159. lecun, 1990, Optimal brain damage, Proc Adv Neural Inf Process Syst, 598
  160. han, 2015, Learning both weights and connections for efficient neural network, Proc Adv Neural Inf Process Syst, 1135
  161. guennebaud, 2015, Eigen V3
  162. denton, 2014, Exploiting linear structure within convolutional networks for efficient evaluation, Proc Adv Neural Inf Process Syst, 1269
  163. 10.1109/MICRO.2014.58 · DOI · Paper page
  164. 10.1109/TC.2016.2574353 · DOI · Paper page
  165. lecun, 1998, The MNIST Database of Handwritten Digits
  166. deshpande, 2018, A Beginner's Guide To Understanding Convolutional Neural Networks
  167. dayhoff, 1990, Neural Network Architectures An Introduction
  168. lecun, 1995, Convolutional networks for images, speech, and time series, The Handbook of Brain Theory and Neural Networks, 3361
  169. hauswald, 2015, DjiNN and Tonic: DNN as a service and its implications for future warehouse scale computers, ACM SIGARCH Comput Archit News, 43, 27, 10.1145/2872887.2749472 · DOI
  170. ng, 2015, Beyond short snippets: Deep networks for video classification, Proc IEEE Conf Comput Vis Pattern Recognit, 4694
  171. lecun, 1990, Handwritten digit recognition with a back-propagation network, Proc Adv Neural Inf Process Syst, 396
  172. 10.1109/HPCA.2017.29 · DOI · Paper page
  173. barros, 2014, A multichannel convolutional neural network for hand posture recognition, Proc Int Conf Artif Neural Netw, 403
  174. 10.1145/3007787.3001140 · DOI · Paper page
  175. huang, 2013, Learning deep structured semantic models for web search using clickthrough data, Proc 22nd ACM Int Conf Conf Inf Knowl Manage, 2333
  176. 10.1109/MNNFS.1996.493811 · DOI
  177. 10.1145/2749469.2750389 · DOI · Paper page
  178. 10.1145/2786763.2694358 · DOI
  179. 10.1145/3007787.3001139 · DOI · Paper page
  180. wolf, 2001, Using embedded processors in hardware models of artificial neural networks, Proc VI Simposio brasileiro de Automao Inteligente
  181. nichols, 2002, Feasibility of floating-point arithmetic in FPGA based artificial neural networks, Proc CAINE, 8
  182. 10.1109/DCV.2002.1218747 · DOI
  183. cardells-tormo, 2005, Area-efficient 2D shift-variant convolvers for FPGA-based digital image processing, Proc IEEE Workshop Signal Process Syst Design Implement, 209

Cited by

See all 496

Metadata

Authors
Shawahna, A., Sait, S. M., & El-Maleh, A.
Venue
IEEE Access
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
2019
Volume
7
Pages
7823-7859
ISSN
2169-3536 (ISSN-L)
DOI
10.1109/access.2018.2890150
OpenAlex
W2907463061
Cited by
496
References
29
FWCI
20.81
Type
journal article
Language
en
Open access
gold
Licence
ieeexplore.ieee.org/Xplorehelp/downloads/license-information/OAPA.html (vor)
Funders
King Fahd University of Petroleum and Minerals

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: FPGA-Based Accelerators of Deep Learning Networks for Learning and Classification: A Review. https://app.cochat.ai/papers/10.1109/access.2018.2890150
The paper
Shawahna, A., Sait, S. M., & El-Maleh, A. (2019). FPGA-Based Accelerators of Deep Learning Networks for Learning and Classification: A Review. IEEE Access. https://doi.org/10.1109/access.2018.2890150

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.