Deep Learning With Edge Computing: A Review
The DOI matches its Crossref record; 130 of the 141 references it cites were found in a registry or an index.
Abstract
Deep learning is currently widely used in a variety of applications, including computer vision and natural language processing. End devices, such as smartphones and Internet-of-Things sensors, are generating data that need to be analyzed in real time using deep learning or used to train deep learning models. However, deep learning inference and training require substantial computation resources to run quickly. Edge computing, where a fine mesh of compute nodes are placed close to end devices, is a viable way to meet the high computation and low-latency requirements of deep learning on edge devices and also provides additional benefits in terms of privacy, bandwidth efficiency, and scalability. This paper aims to provide a comprehensive review of the current state of the art at the intersection of deep learning and edge computing. Specifically, it will provide an overview of applications where deep learning is used at the network edge, discuss various approaches for quickly executing deep learning inference across a combination of end devices, edge servers, and the cloud, and describe the methods for training deep learning models across multiple edge devices. It will also discuss open challenges in terms of systems performance, network technologies and management, benchmarks, and privacy. The reader will take away the following concepts from this paper: understanding scenarios where deep learning at the network edge can be useful, understanding common techniques for speeding up deep learning inference and performing distributed training on edge devices, and understanding recent trends and opportunities.
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
“As mentioned in Section 2.1, the progress of DL is very fast but when it comes to application in Edge or IoT devices then a huge gap is noticeable [67].”
From A survey on deep transfer learning to edge computing for mitigating the COVID-19 pandemic (2020) · cites this paper as [67]
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
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
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
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
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
Going 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
(2016) · Cited by 31,590
Research works printed but not linked: 30
- Jeans, 2019, Related’s Hudson Yards: Smart City or Surveillance City?
- AT&T Multi-Access Edge Computing
- Blog, Edge Computing at Chick-Fil-A
- Goodfellow, 2016, Deep Learning, 1
- 10.1145/3299710.3211336 · DOI · Paper page
- Tensorflow
- Tensorflowlite
- Caffe2
- Caffe
- Pytorch
- Cuda
- Cudnn
- AWS Deeplens
- Deep Learning for Siri’s Voice: On-Device Deep Mixture Density Networks for Hybrid Unit Selection Synthesis
- Google Duplex: An AI System for Accomplishing Real-World Tasks Over the Phone
- Pagespeed Insights: Improve Server Response Time
- 2017, Hey Siri: An On-Device DNN-Powered Voice Trigger for Apple’s Personal Assistant
- Google’s Pixel Buds are no Match for Professional Interpreters
- Jindal, 2018, Enabling Full Body AR With Mask R-CNN2GO
- 10.1145/3081333.3081360 · DOI · Paper page
- Edge TPU.
- VPU.
- NVIDIA EGX Edge Computing Platform
- Qualcomm Neural Processing SDK for AI
- 10.1109/MNET.2018.1700202 · DOI · Paper page
- Arcore Overview
- Teerapittayanon, 2017, BranchyNet: Fast inference via early exiting from deep neural networks, CoRR, abs/1709.01686, 1 · arXiv
- 10.1145/3195970.3196023 · doi:10.1145/3195970.3196023 · Paper page
- Hsieh, Gaia: Geo-distributed machine learning approaching LAN speeds, 629
- 10.1109/NCA.2017.8171350 · DOI · Paper page
Cited by
Items 401–500 of 1,000
DeepAdaIn-Net: Deep Adaptive Device-Edge Collaborative Inference for Augmented Reality
Wang, L., Wu, X., Zhang, Y., Zhang, X., Xu, L., Wu, Z., & Fei, A. (2023) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2023.3312914 · Cited by 23
“With regard to these limitations, edge computing, as a cutting-edge solution, entails delegating computational tasks from the central cloud to individual local edge nodes, leading to a significantly curtailed data transmission latency and the boost in data processing efficiency [6], [7].”
BackgroundA Hybrid Approach for Home Energy Management With Imitation Learning and Online Optimization
Gao, S., Lee, R., Huang, Z.-H., Xiang, C., Yu, M., Tan, K. T., & Lee, T. H. (2023) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2023.3324939 · Cited by 17
“The recent review [31] provides more information.”
Enhancement of edge-based surveillance videos based on bilateral filtering
Chen, S., Xu, J., & Wu, Q. (2024) · Computer Communications · doi:10.1016/j.comcom.2024.01.031 · Cited by 17
Nie, J., Jiang, J., Li, Y., Li, J., Qiao, Y., & Erċışlı, S. (2024) · Internet of Things · doi:10.1016/j.iot.2024.101143 · Cited by 19
Neural Network-based Pattern Recognition in the Framework of Edge Computing
Ning, J. (2024) · Romanian Journal of Information Science and Technology · doi:10.59277/romjist.2024.1.08 · Cited by 18
Chung, K. T., Lee, C. K. M., Tsang, Y. P., Wu, C. H., & Asadipour, A. (2024) · Information Sciences · doi:10.1016/j.ins.2024.121265 · Cited by 20
Adaptive On-Device Model Update for Responsive Video Analytics in Adverse Environments
Kong, Y.-X., Yang, P., & Cheng, Y. (2025) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2024.3466524 · Cited by 20
“Compared with cloud-based solutions [18], edge computing facilities are available in the proximity of source video cameras [19], leading to reduced transmission bandwidth consumption and lower response latency [20].”
Performance Analysis of Deep Neural Networks for Object Classification with Edge TPU
Jainuddin, A. A. A., Hou, Y. C., Baharuddin, M. Z., & Yussof, S. (2020) · International Conference on Information Technology and Multimedia (ICIMU) · doi:10.1109/icimu49871.2020.9243367 · Cited by 17
“These DNN being deployed of course still needs to be prune for use with edge devices, however, depending on the application, these architecture laid out by these researches could potentially help save costs and provide a much more robust application of DNN [13].”
MethodHardware-Aware Affordance Detection for Application in Portable Embedded Systems
Ragusa, E., Gianoglio, C., Dosen, S. S., & Gastaldo, P. (2021) · IEEE Access · doi:10.1109/access.2021.3109733 · Cited by 18
Korol, G., Jordan, M. G., Rutzig, M. B., & Beck, A. C. S. (2021) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3476990 · Cited by 19
“In this scenario, numerous IoT devices (end nodes) send their raw data over local network to Edge servers (sometimes called IoT gateways) that are physically close [7], avoiding the long latency of cloud offloading and increasing security with less data exposure.”
BackgroundTowards Performance Clarity of Edge Video AnalyticsPreprint
Xiao, Z., Xia, Z., Zheng, H.-T., Zhao, B. Y., & Jiang, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.08694 · Cited by 19
“, compute cost to run inference tasks and/or bandwidth cost to transfer data from cameras to servers [29, 74].”
BackgroundPääkkönen, P., Pakkala, D., Kiljander, J., & Sarala, R. (2020) · Future Internet · doi:10.3390/fi13010005 · Cited by 16
“Also, many choices have been explored for increasing the performance of Machine learning (ML)-based inference in cloud-edge continuum [12].”
BackgroundMethodInfluentialJoint Optimization of Model Deployment for Freshness-Sensitive Task Assignment in Edge Intelligence
Liu, H., Liu, S., Long, S., Deng, Q., & Li, Z. (2024) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom52122.2024.10621314 · Cited by 25
Designing on-board explainable passenger flow prediction
Barbareschi, M., Emmanuele, A., Mazzocca, N., & di Torrepadula, F. R. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109648 · Cited by 21
From Technical Prerequisites to Improved Care: Distributed Edge AI for Tomographic Imaging
Akdemir, B., Shahid, H. F., Brix, M. A. K., Lääkkölä, J., Islam, J., Kumar, T., Reponen, J., Nieminen, M. T., & Harjula, E. (2025) · IEEE Access · doi:10.1109/access.2025.3530297 · Cited by 27
Towards scalable memristive hardware for spiking neural networks
Chen, P., Zhang, B., He, E., Xiao, Y., Liu, F., Lin, P., Wang, Z., & Pan, G. (2025) · Materials Horizons · doi:10.1039/d4mh01676a · Cited by 19
EdgeDRNN: Enabling Low-latency Recurrent Neural Network Edge Inference
Gao, C., Ríos Navarro, J. A., Chen, X., Delbrück, T., & Liu, S. (2020) · Deposito de Investigacion Universidad de Sevilla (University of Seville) · Cited by 16
Yoon, G., Kim, G.-Y., Yoo, H., Kim, S., & Kim, R. (2021) · IEEE Access · doi:10.1109/access.2021.3118731 · Cited by 18
“To address this certain drawback, the edge computing paradigm has been employed in object detection offloading strategies [7].”
MethodJoint Social-Aware and Mobility-Aware Computation Offloading in Heterogeneous Mobile Edge Computing
Xu, C., Xu, C. L., Li, B., Li, S., & Li, T. (2022) · IEEE Access · doi:10.1109/access.2022.3158319 · Cited by 18
FDNet: Lightweight Train Image Fault Detection Network in Edge Computing Environments
Zhang, L., Chen, J., Zeng, W., Tong, Z., Wen, Z., & Zhang, X. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3313155 · Cited by 17
“Edge computing can solve the problems of delay, scalability, and privacy of cloud computing [8].”
BackgroundSelf-citationEdgeShield: Enabling Collaborative DDoS Mitigation at the Edge
Xia, X., Chen, F., He, Q., Luo, R., Liu, B., Chua, C., Buyya, R., & Yang, Y. J. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3443260 · Cited by 18
“Edge servers are geographically distributed [7] and remote control from the cloud introduces significant delays in their responses to DDoS attacks [43].”
Rachmanto, R. D., Sukma, Z., Nabhaan, A. N. L., Setyanto, A., Jiang, T., & Kim, I. K. (2024) · IEEE International Conference on Edge Computing and Communications (EDGE) · doi:10.1109/edge62653.2024.00023 · Cited by 25
“The rapid development and significant achievements of deep learning (DL) and artificial intelligence (AI) have led to their widespread adoption in edge computing [1]–[4].”
BackgroundLevisse, A., Bocquet, M., Rios, M. A., Alayan, M., Moreau, M., Nowak, E., Molas, G., Vianello, E., Atienza, D., & Portal, J. (2020) · IEEE Access · doi:10.1109/access.2020.3000867 · Cited by 18
“While new computing architectures and accelerators are being explored to cope with the new computing paradigm opened by machine learning workloads [2], [10], [11], [17]– [19], there is no clear solution on how to store the data needed by the application.”
BackgroundSharing and Caring of Data at the Edge
Trivedi, A., Wang, L., Bal, H. E., & Iosup, A. (2020) · VU Research Portal · Cited by 18
Dynamic Resource Optimization for Adaptive Federated Learning at the Wireless Network Edge
Lorenzo, P. D., Battiloro, C., Merluzzi, M., & Barbarossa, S. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414832 · Cited by 18
Edge-MultiAI: Multi-Tenancy of Latency-Sensitive Deep Learning Applications on Edge
Zobaed, S., Mokhtari, A., Champati, J. P., Kourouma, M., & Salehi, M. A. (2022) · 2022 IEEE/ACM 15th International Conference on Utility and Cloud Computing (UCC) · doi:10.1109/ucc56403.2022.00012 · Cited by 15
“1) The Scope for Edge Intelligence: Numerous research have been undertaken to explore the applications, scopes, and benefits of edge-based AI for the seamless execution of latency-sensitive smart applications [11], [12], [2], [13].”
BackgroundComputer Vision for Site-Specific Weed Management in Precision Agriculture: A Review
Singh, P., Zhao, B., & Shi, Y. (2025) · Agriculture · doi:10.3390/agriculture15212296 · Cited by 20
“A significant obstacle in developing DL models is their typical requirement for extensive datasets during the training phase, with model accuracy generally improving as training image quantities increase [42].”
InfluentialGrad-CAM-Based Classification of Chest X-Ray Images of Pneumonia Patients
Seerala, P. K., & Krishnan, S. (2021) · Communications in computer and information science · doi:10.1007/978-981-16-0425-6_13 · Cited by 15
The gap between technology and agriculture, barrier identification and potential solution analysis
Kasemi, R., Lammer, L., & Vincze, M. (2022) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2022.12.042 · Cited by 17
A Blockchain-based Multi-layer Decentralized Framework for Robust Federated Learning
Wu, D., Wang, N., Zhang, J., Zhang, Y., Xiang, Y., & Gao, L. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9892039 · Cited by 17
“To this end, on-device training or called edge computing could be an effective solution where deep neural network models can be applied on the edge devices or the Internet of things [3].”
BackgroundZeleke, D. A., & Kim, H. (2023) · Aerospace · doi:10.3390/aerospace10010078 · Cited by 20
Programming for High-Performance Computing on Edge Accelerators
Kang, P. (2023) · Mathematics · doi:10.3390/math11041055 · Cited by 14
Kumar, R., Sangwan, K. S., Herrmann, C. A., & Thakur, S. (2023) · International Journal of Computer Integrated Manufacturing · doi:10.1080/0951192x.2023.2189312 · Cited by 18
“Edge computing is an ideal platform for deploying deep learning models featuring high computation, low-latency, and with added advantages such as privacy, bandwidth efficiency, and scalability (Chen and Ran 2019). Manufacturing industries can also benefit in terms of higher efficiency and cost reduction from the imple…”
BackgroundMethodInfluentialHutiri, W. T., Ding, A. Y., Kawsar, F., & Mathur, A. (2023) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3591867 · Cited by 25
“Rising concerns about digital privacy and personal data protection [43] are motivating a shift in data processing and machine learning (ML) from cloud servers to end devices [10].”
BackgroundData-Aware Adaptive Compression for Stream Processing
Zhang, Y., Zhang, F., Li, H., Zhang, S., Guo, X., Chen, Y., Pan, A., & Du, X. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3377710 · Cited by 20
“FollowingtherapidprogressofIn-ternetofThings(IoT)technology,edgecomputinghasexpanded intotherealmoflow-powerandembeddeddevices,suchas RaspberryPi4B[43]andmicrocontrollers[44].”
Feng, Z.-Y., Wang, Z., Zheng, K., Li, R., Zhao, Y., & Wang, Y. (2024) · Ocean Engineering · doi:10.1016/j.oceaneng.2024.117539 · Cited by 16
Decentralized Federated GAN for Hyperspectral Change Detection in Edge Computing
Xie, W., Xu, X., & Li, Y. (2024) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2024.3389641 · Cited by 18
“In a simulated orbital edge computing environment, experimental results indicate that FedGAN reduces communication overhead costs by 10–15 times compared to existing approaches when tested on different models/datasets [18], [19].”
BackgroundChi, T.-K., Chen, T., Lin, Y., Lin, T., Zhang, J., Lu, C., Chen, S., Li, K.-C., & Abu, P. A. R. (2024) · Sensors · doi:10.3390/s24103098 · Cited by 20
“The edge computing devices that perform AI computations locally have been discussed by [3,4].”
BackgroundSecure architecture for Industrial Edge of Things(IEoT): A hierarchical perspective
Li, P., Xia, J., Wang, Q., Zhang, Y., & Wu, M. (2024) · Computer Networks · doi:10.1016/j.comnet.2024.110641 · Cited by 18
The convergence of edge computing and supply chain resilience in retail marketing
Anozie, U. C., Dada, S. A., Okonkwo, F. C., Egunlae, O. O., Animasahun, B. O., & Mazino, O. (2024) · International Journal of Science and Research Archive · doi:10.30574/ijsra.2024.12.2.1574 · Cited by 23
Stragglers-Aware Low-Latency Synchronous Federated Learning via Layer-Wise Model Updates
Lang, N., Cohen, A., & Shlezinger, N. (2024) · IEEE Transactions on Communications · doi:10.1109/tcomm.2024.3486979 · Cited by 25
“To avoid data sharing, FL exploits the local computational capabilities of edge users [6], training locally with periodic aggregations orchestrated by a server.”
MethodA Novel Multi-step Finite-State Automaton for Arbitrarily Deterministic Tsetlin Machine Learning
Abeyrathna, K. D., Granmo, O., Shafik, R., Yakovlev, A., Wheeldon, A., Lei, J., & Goodwin, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-63799-6_8 · Cited by 15
Memory-Latency-Accuracy Trade-Offs for Continual Learning on a RISC-V Extreme-Edge Node
Ravaglia, L., Rusci, M., Capotondi, A., Conti, F., Pellegrini, L., Lomonaco, V., Maltoni, D., & Benini, L. (2020) · SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems · doi:10.1109/sips50750.2020.9195220 · Cited by 19
Evaluation of Deep Learning Accelerators for Object Detection at the Edge
Puchtler, P., & Peinl, R. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58285-2_29 · Cited by 23
Sliding Differential Evolution Scheduling for Federated Learning in Bandwidth-Limited Networks
Luo, Y., Xu, J., Xu, W., & Wang, K. (2020) · IEEE Communications Letters · doi:10.1109/lcomm.2020.3032517 · Cited by 16
“Staleness method leverages Tk[t] to avoid overtraining and under-training [11].”
MethodOptimization of Deep Learning Inference on Edge Devices
Kristiani, E., Yang, C., & Nguyen, K. L. P. (2020) · International Conference on Pervasive Artificial Intelligence (ICPAI) · doi:10.1109/icpai51961.2020.00056 · Cited by 21
“Chen et al. [6] aimed to provide a systematic analysis at the intersection of deep learning and edge computing of the existing state of the art.”
BackgroundEdge computing: A systematic mapping study
Sakhdari, J., Zolfaghari, B., Izadpanah, S., Zargar, S. M., Rahati-Quchani, M., Shadi, M., Abrishami, S., & Rasoolzadegan, A. (2023) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.7741 · Cited by 20
Two-Dimensional Task Offloading for Mobile Networks: An Imitation Learning Framework
Yan, Z., Cheng, P., Chen, Z., Vucetic, B. S., & Li, Y. (2021) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2021.3093452 · Cited by 18
Low-precision Logarithmic Number Systems: Beyond Base-2
Alam, S. A., Garland, J. W., & Gregg, D. (2021) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3461699 · Cited by 20
“Sending data back to the cloud for processing requires large amounts of energy, and may become a bottleneck that fails to meet latency requirements for real-time embedded and autonomous systems while also stretching the limited processing, memory and/or battery capacity of embedded devices [12, 36].”
BackgroundCompact CNN Models for On-device Ocular-based User Recognition in Mobile Devices
Almadan, A., & Rattani, A. (2021) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci50451.2021.9660033 · Cited by 17
Adaptive and Collaborative Edge Inference in Task Stream with Latency Constraint
Song, J., Liu, Z., Wang, X., Qiu, C., & Chen, X. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500892 · Cited by 19
Resource Allocation Using Deep Learning in Mobile Small Cell Networks
Zafar, S., Jangsher, S., & Al‐Dweik, A. (2022) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2022.3146487 · Cited by 15
Material measurement units for a circular economy: Foundations through a review
Zocco, F., McLoone, S., & Smyth, B. M. (2022) · Sustainable Production and Consumption · doi:10.1016/j.spc.2022.05.022 · Cited by 17
Edge Intelligence with Distributed Processing of DNNs: A Survey
Tang, S., Cui, M., Qi, L., & Xu, X. (2022) · Computer Modeling in Engineering & Sciences · doi:10.32604/cmes.2023.023684 · Cited by 20
Memristor Crossbar Circuits Implementing Equilibrium Propagation for On-Device Learning
Oh, S., An, J., Cho, S., Yoon, R., & Min, K. (2023) · Micromachines · doi:10.3390/mi14071367 · Cited by 24
A Blockchain-Assisted Distributed Edge Intelligence for Privacy-Preserving Vehicular Networks
Firdaus, M., Larasati, H. T., & Rhee, K.-H. (2023) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2023.039487 · Cited by 20
Personalized Dietary Recommendations Using Machine Learning: A Comprehensive Review
Varshney, N., Jadhav, N., Gupta, K., Mate, N. R., Rose, A., & Kumar, P. (2023) · International Conference on Artificial Intelligence for Innovations in Healthcare Industries (ICAIIHI) · doi:10.1109/icaiihi57871.2023.10489126 · Cited by 18
“For instance, people who carried particular metabolic gene markers responded differently to dietary suggestions that were specifically suited to them [24].”
BackgroundPrune-FSL: Pruning-Based Lightweight Few-Shot Learning for Plant Disease Identification
Yan, W., Feng, Q., Yang, S., Zhang, J., & Yang, W. (2024) · Agronomy · doi:10.3390/agronomy14091878 · Cited by 14
A Deep Learning-Based Car Accident Detection Framework Using Edge and Cloud Computing
Banerjee, S., Mondal, M. K., Roy, M., Alnumay, W. S., & Biswas, U. (2024) · IEEE Access · doi:10.1109/access.2024.3458420 · Cited by 16
FLEdge: Benchmarking Federated Learning Applications in Edge Computing Systems
Woisetschläger, H., Erben, A., Mayer, R., Wang, S., & Jacobsen, H. (2024) · International Middleware Conference · doi:10.1145/3652892.3700751 · Cited by 27
autoGEMM: Pushing the Limits of Irregular Matrix Multiplication on Arm Architectures
Wu, D., Meng, J., Zhu, W., Deng, M., Wang, X., Luo, T., Wahib, M., & Wei, Y. (2024) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1109/sc41406.2024.00027 · Cited by 20
“This can have a significant impact on the overall efficiency of various DL applications [18], [19], in both edge devices and data centers.”
Inverse Design of FBG-Based Optical Filters Using Deep Learning: A Hybrid CNN-MLP Approach
Adibnia, E., Ghadrdan, M., & Mansouri-Birjandi, M. A. (2025) · Journal of Lightwave Technology · doi:10.1109/jlt.2025.3534275 · Cited by 18
“We are also cognizant of the propensity for model overfitting [29], the limitations in model interpretability [30], and the significant computational requirements inherent in the deployment of DL architectures [31].”
Machine Learning at the Edge: Efficient Utilization of Limited CPU/GPU Resources by Multiplexing
Dhakal, A., Kulkarni, S. G., & Ramakrishnan, K. K. (2020) · IEEE International Conference on Network Protocols (ICNP) · doi:10.1109/icnp49622.2020.9259361 · Cited by 14
“Edge clouds are often used to offload processing from end-devices [5]–[7].”
BackgroundModel-based Weight Quantization for Convolutional Neural Network Compression
Gheorghe, Ș., & Ivanovici, M. (2021) · International Conference on Engineering of Modern Electric Systems · doi:10.1109/emes52337.2021.9484143 · Cited by 14
A collaborative deep learning microservice for backdoor defenses in Industrial IoT networks
Liu, Q., Chen, L., Jiang, H., Wu, J., Wang, T., Peng, T., & Wang, G. (2021) · Ad Hoc Networks · doi:10.1016/j.adhoc.2021.102727 · Cited by 16
“Deep neural networks (DNNs) demonstrate the distinguished capabilities in discovering high-dimensional structures from massive volumes of data and have been applied to a wide range of IIoT fields, such as face recognition, disease diagnosis, autonomous driving and so on [1,2].”
BackgroundXu, Y., dos Santos, M. A., Souza, L. F. D. F., Marques, A. G., Zhang, L., Nascimento, J. J. D. C., de Albuquerque, V. H. C., & Filho, P. P. R. (2022) · IET Image Processing · doi:10.1049/ipr2.12449 · Cited by 17
Zhao, J., Cao, B., Liu, X., Yang, P., Singh, A. K., & Lv, Z. (2022) · IEEE Transactions on Fuzzy Systems · doi:10.1109/tfuzz.2022.3163909 · Cited by 15
“DNN has been applied to some difficult practical problems [18].”
MethodResilience Enhancement at Edge Cloud Systems
Moura, J., & Hutchison, D. (2022) · IEEE Access · doi:10.1109/access.2022.3165744 · Cited by 16
“moved from remote clouds [10] to edge clouds [11] in order to diminish the data and service access latency, to provide the edge network infrastructure with local scalable processing, and even to run local self-adaptable algorithms [12].”
BackgroundECCVideo: A Scalable Edge Cloud Collaborative Video Analysis System
Han, Q., Ren, X., Zhao, P. F., Wang, Y., Wang, L., Zhao, C., & Yang, X. (2022) · IEEE Intelligent Systems · doi:10.1109/mis.2022.3214614 · Cited by 14
Towards Data-Efficient Continuous Learning for Edge Video Analytics via Smart Caching
Zhang, L., Gao, G., & Zhang, H. (2022) · ACM Conference on Embedded Networked Sensor Systems (SenSys) · doi:10.1145/3560905.3568430 · Cited by 15
“The edge computing paradigm is widely adopted for video analytics applications due to the benefits of low data transmission delays and better data privacy protection [7, 16, 30].”
MERRC: A Memristor-Enabled Reconfigurable Low-Power Reservoir Computing Architecture at the Edge
Nowshin, F., Huang, Y., Sarkar, M. R., Xia, Q., & Yi, Y. (2024) · IEEE Transactions on Circuits and Systems I: Regular Papers · doi:10.1109/tcsi.2023.3329337 · Cited by 12
“When deployed on edge devices, deep learning becomes exceptionally advantageous, especially in domains like natural language processing, computer vision, and the IoT [12], [13].”
BackgroundYu, T., Yao, S., Zhang, Z., Ramos, A. F. V., RODRIGUEZ‐SAONA, L. E., & Wang, J. (2025) · Trends in Food Science & Technology · doi:10.1016/j.tifs.2025.105017 · Cited by 13
The design of analogue in-memory computing tiles
Singh, A., Gallo, M. L., Vasilopoulos, A., Luquin, J., Narayanan, P., Burr, G. W., & Sebastian, A. (2025) · Nature Electronics · doi:10.1038/s41928-025-01537-5 · Cited by 21
“(See Supplementary Notes 2 and 3 for further details of the noise models and a component-wise analysis.)”
Yoon, J., Byeon, Y., Kim, J., & Lee, H. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3131407 · Cited by 19
“Edge devices tend to be resource constrained and communication-fragile due to the low-power wireless properties of transmission and reception, so the edge computing environment makes parallelism in deep learning challenging [16], [17].”
BackgroundPerformance Modeling of Computer Vision-based CNN on Edge GPUs
Bouzidi, H., Ouarnoughi, H., Niar, S., & Cadi, A. A. E. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3527169 · Cited by 19
“Finding the best combination between CNNs models and HW platforms is a tedious task [12] that needs a long time to explore the large and complex design space.”
BackgroundAnomaly Detection in Coastal Wireless Sensors via Efficient Deep Sequential Learning
Matar, M., Xia, T., Huguenard, K. D., Huston, D. R., & Wshah, S. (2023) · IEEE Access · doi:10.1109/access.2023.3322370 · Cited by 18
Lightweight Detection Model RM-LFPN-YOLO for Rebar Counting
Liu, H., Cheng, W., Li, C.-W., Xu, Y., & Fan, S. (2024) · IEEE Access · doi:10.1109/access.2024.3349978 · Cited by 13
Makondo, N., Kobo, H. I., Mathonsi, T. E., & Mamushiane, L. (2023) · International Conference on Electrical, Computer and Energy Technologies (ICECET) · doi:10.1109/icecet58911.2023.10389349 · Cited by 17
“It recognizes users based on their Internet Protocol (IP) addresses and offloads user traffic in accordance with the traffic offloading rules [37].”
BackgroundQian, C., Ling, T., & Schiele, G. (2024) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2024.103181 · Cited by 15
Private Over-the-Air Federated Learning at Band-Limited Edge
Tao, Y., Chen, S., Zhang, C., Wang, D., Yu, D., Cheng, X., & Dressler, F. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3411295 · Cited by 20
“To leverage the potential of the edge Big Data, it is imperative to implement large-scale machine learning algorithms at the edge [1], [2], [3], which fosters the concept of edge intelligence and facilitates various innovative applications that improve human welfare, such as smart cities [4], [5],…”
BackgroundResource-efficient Parallel Split Learning in Heterogeneous Edge Computing
Zhang, M., Cao, J., Sahni, Y., Chen, X., & Jiang, S. (2024) · 2012 International Conference on Computing, Networking and Communications (ICNC) · doi:10.1109/icnc59896.2024.10556386 · Cited by 19
“To address these challenges, a new paradigm known as Edge AI has emerged, emphasizing the training and implementation of AI models on edge devices (like edge servers, gateways, and smartphones), which are situated closer to where the data originates [2].”
Moitra, A., Bhattacharjee, A., Li, Y., Kim, Y., & Panda, P. (2024) · Applied Physics Reviews · doi:10.1063/5.0211040 · Cited by 19
“Furthermore, for applications requiring real-time or near-real-time responses, such as autonomous vehicles or emergency response systems, online learning can save huge communication latency and bandwidth as data does not need to be communicated back and forth across the internet 151,152 .”
BackgroundBeamforming and Phased Arrays: Elevating MIMO-OFDM Systems for Next-Generation Wireless Networks
Upadhyay, D., Sharma, K. B., Gupta, M., Upadhyay, A. B., & Yadav, S. P. (2024) · International Conference on Electrical Electronics and Computing Technologies (ICEECT) · doi:10.1109/iceect61758.2024.10739319 · Cited by 13
“These systems must evolve into ever-more sophisticated designs with more antenna components and allocated resources, such as subcarriers [2], because to the relentless need for higher data rates.”
BackgroundSignEdgeLVM transformer model for enhanced sign language translation on edge devices
Damdoo, R., & Kumar, P. (2025) · Discover Computing · doi:10.1007/s10791-025-09509-1 · Cited by 19
“Chen et al. [42] review how deep learning can be integrated with edge computing to reduce latency, improve privacy, and enable adaptable real-time applications.”
Rokai, J., Rácz, M., Fiáth, R., Ulbert, I., & Márton, G. (2020) · Journal of Neural Engineering · doi:10.1088/1741-2552/abf521 · Cited by 12
“Modern artificial intelligence accelerator application-specific integrated circuits such as tensor processing units (TPUs) [9] allow the inference of deep convolutional neural networks (CNNs) on wearable devices.”
RCT: Resource Constrained Training for Edge AI
Huang, T., Luo, T., Yan, M., Zhou, J. T., & Goh, R. S. M. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3190451 · Cited by 19
“E DGE computing emerges as an attractive alternative to cloud computing, for its advantages in data privacy, response latency, and energy saving for data transmissions [1].”
BackgroundHashash, O., Sharafeddine, S., & Dawy, Z. (2021) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops50388.2021.9473646 · Cited by 14
Soudain: Online Adaptive Profile Configuration for Real-time Video Analytics
Wu, K., Jin, Y., Miao, W., Zeng, Z., Qian, Z., Wang, J., Zhou, M., & Cao, T. (2021) · International Workshop on Quality of Service · doi:10.1109/iwqos52092.2021.9521328 · Cited by 17
A Review on Deep Reinforcement Learning for the management of SDN and NFV in Edge-IoT
Alonso, R. S., Prieto, J., la Prieta, F. D., Rodríguez, S., & Corchado, J. M. (2021) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps52748.2021.9682179 · Cited by 16
“However, the work of [14] does not mention possible applications based on DRL. Reference [15] present a detailed analysis of the application of DL solutions in the field of EC.”
BackgroundJoint Privacy Enhancement and Quantization in Federated Learning
Lang, N., & Shlezinger, N. (2022) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit50566.2022.9834551 · Cited by 16
Joint Task Offloading, CNN Layer Scheduling, and Resource Allocation in Cooperative Computing System
Chai, R., Song, X., & Chen, Q. (2020) · IEEE Systems Journal · doi:10.1109/jsyst.2020.2991814 · Cited by 14
“As a typical CNN is composed of multiple layers with each layer having various data processing capability, it is possible to schedule a number of layers of the CNNs, process a portion of a task at one computing server, and execute the remaining task at other servers [4], [5].”
BackgroundDistributed perception and model inference with intelligent connected vehicles in smart cities
Li, C., Wang, S., Li, X., Zhao, F., & Yu, R. (2020) · Ad Hoc Networks · doi:10.1016/j.adhoc.2020.102152 · Cited by 14
Understanding the Threats of Trojaned Quantized Neural Network in Model Supply Chains
Pan, X., Zhang, M., Yan, Y., & Yang, M. (2021) · Annual Computer Security Applications Conference · doi:10.1145/3485832.3485881 · Cited by 23
“With the ever-evolving embedded system hardware, edge devices including smartphones and smart house devices are gathering and generating more informative data than ever, which gives birth to the trend of deploying well-trained deep neural networks (DNN) on edge devices to enable better analytic and inference capabilit…”
BackgroundA Data and Compute Efficient Design for Limited-Resources Deep LearningPreprint
Mohamed, M., Cesa, G., Cohen, T., & Welling, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.09691 · Cited by 29
“Deep learning has recently moved closer to edge devices Chen & Ran (2019), creating opportunities for many new applications where data needs to be analyzed in real time.”
BackgroundKing, T., Zhou, Y., Röddiger, T., & Beigl, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-90764-z · Cited by 24
Radar Perception for Autonomous Unmanned Aerial Vehicles: a Survey
Corradi, F., & Fioranelli, F. (2022) · doi:10.1145/3522784.3522787 · Cited by 14
“Because scaling up on-device intelligence by scaling sensor resolution, model size, and computing needs is not a viable solution on edge devices, much research is devoted to developing devices capable of accelerating the inference of deep learning workloads [6].”
BackgroundArtificial intelligence and digital twins in power systems: Trends, synergies and opportunities
Shen, Z., Arraño-Vargas, F., & Konstantinou, G. S. (2022) · Digital Twin · doi:10.12688/digitaltwin.17632.1 · Cited by 60
Amogi, B. R., Ranjan, R., & Khot, L. R. (2023) · Information Processing in Agriculture · doi:10.1016/j.inpa.2023.12.001 · Cited by 21
Load-aware task migration algorithm toward adaptive load balancing in Edge Computing
Zhu, X., Yao, W., & Wang, W. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.03.014 · Cited by 18
An Autonomous Parallelization of Transformer Model Inference on Heterogeneous Edge Devices
Lee, J., Bahk, I., Kim, H., Jeong, S., Lee, S., & Min, D. (2024) · International Conference on Supercomputing · doi:10.1145/3650200.3656628 · Cited by 20
“Unlike the traditional AI workflow, which sends data to the cloud, Edge AI benefits from low latency and low network traffic [4, 13, 21, 42].”
BackgroundMethod
Metadata
- Authors
- Chen, J., & Ran, X.
- Affiliations
- University of California, Riverside (US)
- ORCID
- Chen, Jiasi
- Venue
- Proceedings of the IEEE
- Published
- 2019-08
- Volume
- 107
- Issue
- 8
- Pages
- 1655-1674
- ISSN
- 0018-9219 (ISSN-L)
- 1558-2256
- OpenAlex
- W2960833983
- Cited by
- 1,540
- References
- 141
- Citations by year
- 2026: 205 · 2025: 250 · 2024: 270 · 2023: 245 · 2022: 232 · 2021: 218 · 2020: 114 · 2019: 6
- FWCI
- 77.78
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- Proceedings of the IEEE (publishedVersion)
- Funders
- National Science Foundation (CNS-1817216)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Information Systems › Internet of Things and AI
- Keywords
- edge computing, deep learning, edge inference, distributed deep learning training, edge servers, cloud-edge collaboration, Internet of Things, low latency, privacy, spectral efficiency
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.