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 701–800 of 1,000
Modelling multi-cell edge video analytics
Peris, J. A., & Fodor, V. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838556 · Cited by 6
“At the same time, we have experienced major breakthroughs in deep learning, which have made it possible to manage and leverage the rise in data volumes for solving complex problems, such as image classification, language processing, or intrusion detection [1].”
BackgroundA Framework and Method for Surface Floating Object Detection Based on 6G Networks
Li, H., Yang, S., Liu, J., Yang, Y., Kadoch, M., & Liu, T. (2022) · Electronics · doi:10.3390/electronics11182939 · Cited by 7
Computational Optimization of Image-Based Reinforcement Learning for Robotics
Ferraro, S., de Maele, T. V., Mazzaglia, P., Verbelen, T., & Dhoedt, B. (2022) · Sensors · doi:10.3390/s22197382 · Cited by 10
“Deployability on an edge device is of major interest for industrial applications where resources are often limited and where security issues arise from the upload of the raw sensory information to the cloud [16,17].”
VEI: a multicloud edge gateway for computer vision in IoT
Luu, S., Ravindran, A., Pazho, A. D., & Tabkhi, H. (2022) · doi:10.1145/3565385.3565877 · Cited by 7
“Other efforts have focused on edge friendly implementation of deep learning models [4].”
BackgroundBenchmarking Container Technologies For IoT Environments
Jing, Y., Qiao, Z., & Sinnott, R. (2022) · International Conference on Fog and Mobile Edge Computing (FMEC) · doi:10.1109/fmec57183.2022.10062773 · Cited by 7
“It’s goal is to be energyefficient, minimise bandwidth, and offer low-latency features that complement the problems encountered in centralized computing models [4].”
BackgroundAn Event-Classification Neural Network Approach for Rapid Railroad Bridge Impact Detection
Lawal, O., Shajihan, S. A. V., Mechitov, K. A., & Spencer, B. F. (2023) · Sensors · doi:10.3390/s23063330 · Cited by 7
“Edge computing has been proposed by researchers as a feasible solution to this problem [47].”
BackgroundObfuscating the Dataset: Impacts and Applications
Yu, G., Wang, X., Sun, C., Yu, P., Ni, W., & Liu, R. P. (2023) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3597936 · Cited by 9
“1 INTRODUCTION Artiicial Intelligence (AI) has evolved for decades and the capability of machine learning has become one of the core components in many applications in human daily life, including edge computing networks [1, 2].”
BackgroundFederated Blockchain Learning at the Edge
Calo, J., & Lo, B. P. L. (2023) · Information · doi:10.3390/info14060318 · Cited by 6
“On the other hand, IoT devices using machine learning do exist and bringing machine learning, particularly neural networks and their collective family (convolutional, deep, etc.), onto the edge is a wide area of research [12].”
BackgroundHu, W., Zhou, Y., Lu, J., Sheng, J., & Jin, Z. (2023) · Ecological Chemistry and Engineering S · doi:10.2478/eces-2023-0029 · Cited by 9
McDonnell, K. J. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12144830 · Cited by 12
“A complementary data analytic approach, edge computing, redistributes data processing, computations and memory storage from HPCC hubs to smaller, local computer nodes contiguous with the data source [55].”
BackgroundArtificial Intelligence Integrated with Machine Learning for Enhancing Business in E-World
Kiran, K. R., Kumar, B. H., Mishra, B. S., Yadav, A. S., Nagendar, Y., & Tiwari, M. (2023) · International Conference on Inventive Research in Computing Applications (ICIRCA) · doi:10.1109/icirca57980.2023.10220756 · Cited by 7
“According to the recent survey, machine learning is used by 67% of companies [11-15].”
Toward Improving Ensemble-Based Collaborative Inference at the Edge
Kumazawa, S., Yu, J., Kawamura, K., Chu, T. V., & Motomura, M. (2024) · IEEE Access · doi:10.1109/access.2024.3351308 · Cited by 8
Liu, R., & Wang, Y. (2024) · Journal of the Knowledge Economy · doi:10.1007/s13132-024-02087-5 · Cited by 14
“The conclusion is presented in the “Conclusion” section.”
Retracted3D Point Cloud Object Detection on Edge Devices for Split Computing
Noguchi, T., & Azumi, T. (2024) · IEEE Real-Time and Intelligent Edge Computing Workshop (RAGE) · doi:10.1109/rage62451.2024.00009 · Cited by 7
“To address this problem, a method using edge servers with higher computational power has been proposed [7].”
An Overview of the Edge Computing in the Modern Digital Age
França, R. P., Monteiro, A. C. B., Arthur, R., & Iano, Y. (2021) · Advances in information security · doi:10.1007/978-3-030-57328-7_2 · Cited by 8
Veronesi, A., Krstic, M. D., & Bertozzi, D. (2020) · IFIP/IEEE International Conference on Very Large Scale Integration · doi:10.1109/vlsi-soc46417.2020.9344109 · Cited by 6
“Designing efficient hardware systems to support deep learning is an important step towards enabling its wide deployment, particularly for embedded applications such as mobile, Internet of Things (IoT), and drones [1], [2].”
BackgroundGušev, M. (2022) · Elsevier eBooks · doi:10.1016/b978-0-12-824054-0.00020-4 · Cited by 6
Towards supervised real-time human activity recognition on embedded equipment
Najeh, H., Lohr, C., & Leduc, B. (2022) · IEEE International Workshop on Metrology for Living Environment (MetroLivEnv) · doi:10.1109/metrolivenv54405.2022.9826937 · Cited by 9
“1) Deep Learning Benchmarks on Edge Devices: An important number of research works has addressed deep learning operating at the edge of the network [37].”
BackgroundRemoving Channel Estimation by Location-Only Based Deep Learning for RIS Aided Mobile Edge Computing
Hu, X., Masouros, C., & Wong, K. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500961 · Cited by 6
“Note that deep learning methods have been investigated in several MEC systems to simplify the optimization algorithm [17–19].”
MethodEdge AI for Real-Time Video Analytics in Surveillance Systems
Sree, K. P. N. V. S. (2023) · International Journal on Recent and Innovation Trends in Computing and Communication · doi:10.17762/ijritcc.v11i10.8947 · Cited by 5
Model-Distributed Inference for Large Language Models at the Edge
Macario, D., Seferoğlu, H., & Koyuncu, E. (2025) · IEEE International Symposium on Local and Metropolitan Area Networks (LANMAN) · doi:10.1109/lanman66415.2025.11154542 · Cited by 17
“Thus, many approaches have been devised in the literature for performing traditional deep neural network inference at the edge [5], [4].”
Spiking neural networks for EEG signal analysis: From theory to practice
Cai, S., Lin, Z., Liu, X., Wei, W., Wang, S., Zhang, M., Schultz, T., & Li, H. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.108127 · Cited by 12
Feng, Y. (2025) · Scientific Reports · doi:10.1038/s41598-025-18344-9 · Cited by 14
“The edge computing implementation significantly reduces processing delays and enables deterministic real-time performance that supports precise collaborative coordination among multiple robotic units 56 .”
Kang, P. (2025) · Mathematics · doi:10.3390/math13203305 · Cited by 7
Deep Learning on RISC-V Platforms at the Edge: A Perspective on the Hardware and Software Support
Agosta, G., Galimberti, A., & Zoni, D. (2025) · ACM Computing Surveys · doi:10.1145/3772277 · Cited by 15
“Second, a large body of surveys addresses DL and ML at the edge more broadly, covering TinyML [22], distributed training and inference [28], deployment on MCUs and smartphones [19, 182], and tools and frameworks for edge ML [121].”
An in-depth review and analysis of deep learning methods and applications in spinal cord imaging
Hossain, M. S., Rahman, M., Ahmed, M., Rahman, A., Kabir, M. M., Mridha, M. F., & Shin, J. (2025) · Healthcare Analytics · doi:10.1016/j.health.2025.100429 · Cited by 8
DNN Partitioning for Cooperative Inference in Edge Intelligence: Modeling, Solutions, Toolchains
Hao, Y., Ding, N., Xia, W., Ge, H., & Xu, L. (2026) · ACM Computing Surveys · doi:10.1145/3786145 · Cited by 14
“…Y. Hao et al. [34] Cloud&Edge&End [35] Device&Server [36] – [37] Cloud&Edge&End [38] Device-server [39] – [40] Cloud&Edge&End [41] Cloud&Edge&End [42] Edge&End [43] Device&Server [44] – [45] Cloud&Edge&End Thispaper Cloud&Edge&End task offloading, and model optimization are tightly interwoven…”
Fernandes, T. B., Fernandes, E. C., & Borsato, M. (2026) · Journal of the Brazilian Society of Mechanical Sciences and Engineering · doi:10.1007/s40430-026-06408-4 · Cited by 8
Application and Implementation of Deep Learning for Evaluation of Martial Arts Trainings
Jin, M. (2022) · Mobile Information Systems · doi:10.1155/2022/3979817 · Cited by 6
“…was published in “Science” in 2006, deep learning has quickly become a research hotspot and has been widely used in image classification, face recognition, speech recognition, artificial intelligence, etc. [18]. application of deep learning in the evaluation of martial arts instruction is suitable.”
BackgroundConfAx: Exploiting Approximate Computing for Configurable FPGA CNN Acceleration at the Edge
Korol, G., Jordan, M. G., Rutzig, M. B., & Beck, A. C. S. (2022) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas48785.2022.9937676 · Cited by 7
“In this scenario, numerous IoT devices (end nodes) send their raw data over the local network to be processed by Edge servers (or IoT gateways) that are physically close [1], avoiding the long latency and increased risk of Cloud offloading.”
BackgroundTwo-Step Knowledge Distillation for Tiny Speech Enhancement
Nathoo, R. D., Kegler, M., & Stamenovic, M. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446796 · Cited by 10
“Pruning is a popular method for reducing the size of DNN models for SE [4, 5, 6, 8].”
BackgroundMethodInfluentialNS-Engine: Near-Sensor Neural Network Engine with SRAM-Based Compute-in-Memory Macro
Ren, E., Liu, J., Luo, L. Z., Yang, X., Wei, Q., & Qiao, F. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10557977 · Cited by 5
“In recent years, Internet of Things (IoT) technology has developed rapidly, and computing nodes have partially migrated from the cloud to the edge due to bandwidth, power, and security considerations [1].”
BackgroundSecure and Efficient Face Recognition via Supervised Federated Learning
Solomon, E., Woubie, A., Emiru, E. S., & Abdelzaher, A. F. (2024) · IEEE RO-MAN · doi:10.1109/ro-man60168.2024.10731228 · Cited by 5
“These systems typically utilize ML techniques to train deep neural networks, primarily gathered from end-devices like smartphones, with model training occurring on centralized servers [2].”
BackgroundLing, Z., Jiang, X., Tan, X., He, H., Zhu, S., & Yang, J. (2024) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2024.3506588 · Cited by 6
“The first involves scenarios such as edge computing [38] and IoT [39], where data is collected by sensors and transmitted to the training system for processing.”
A design of magnetic tunnel junctions for the deployment of neuromorphic hardware for edge computing
Rodrigues, D. R., Raimondo, E., Tomasello, R., Carpentieri, M., & Finocchio, G. (2025) · Applied Physics Letters · doi:10.1063/5.0237090 · Cited by 7
Upadhyay, D., Sharma, K. B., Gupta, M., Upadhyay, A. B., & Yadav, S. P. (2024) · International Conference on Communication, Computing and Energy Efficient Technologies (I3CEET) · doi:10.1109/i3ceet61722.2024.10993891 · Cited by 5
“The Convolutional neural networks (CNNs), which have been developed [2] and refined to provide powerful tools for image identification, classification, and analysis, have revolutionised the science of computer vision.”
Ye, S., Yu, F., Hao, M., Wu, M., Chen, D., Bu, C., Lu, P., Du, Y., & Cui, K. (2025) · International Journal of Energy Research · doi:10.1155/er/8824853 · Cited by 5
“Additionally, lightweight models can be developed for edge computing applications [52].”
Edge accelerated reconstruction using sensitivity analysis for single-lens computational imaging
Wang, X., Feng, T., Xing, Y., Zhao, Z., Dun, X., Wang, Z., & Cheng, X. (2025) · Advanced imaging. · doi:10.3788/ai.2025.10003 · Cited by 6
Rahman, M. J., & Morshed, B. I. (2021) · ACM Transactions on Computing for Healthcare · doi:10.1145/3479432 · Cited by 6
Fast and Automatic Object Registration for Human-Robot Collaboration in Industrial Manufacturing
Geiß, M., Baresch, M., Chasparis, G. C., Schweiger, E., Teringl, N., & Zwick, M. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-14343-4_22 · Cited by 5
ReLP: Reinforcement Learning Pruning Method Based on Prior Knowledge
Zhang, W., Ji, M., Yu, H., & Zhen, C. (2022) · Neural Processing Letters · doi:10.1007/s11063-022-11058-3 · Cited by 9
“What follows is a huge demand for machine computing power, CNN usually occupies a lot of memory space, and the computing cost is high [1, 2].”
BackgroundMigliorelli, L., Cacciatore, A., Ottaviani, V., Berardini, D., Dellacà, R. L., Frontoni, E., & Moccia, S. (2022) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-022-02696-9 · Cited by 7
“Inspired by [9], to quantitatively evaluate the efficiency of each model presented in Section 4.”
BackgroundMethodInfluentialTowards a Methodology for Building Dynamic Urgent Applications on Continuum Computing Platforms
Balouek‐Thomert, D., Caron, E., Lefèvre, L., & Parashar, M. (2022) · doi:10.1109/ciw-ius56691.2022.00009 · Cited by 9
“[12], [13].”
Over-the-Air Split Learning with MIMO-Based Neural Network and Constellation-Based Activation
Yang, Y.-Z., Zhang, Z.-Y., & Yang, Z.-H. (2022) · International Workshop on Machine Learning for Signal Processing · doi:10.1109/mlsp55214.2022.9943518 · Cited by 9
“Due to its flexible framework over wireless networks, SL has shown the superiority in AI-empowered systems [5].”
BackgroundSpeeding up Smartphone-Based Dew Computing: In Vivo Experiments Setup Via an Evolutionary Algorithm
Yannibelli, V., Hirsch, M., Toloza, J. M., Majchrzak, T. A., Zunino, A., & Mateos, C. (2023) · Sensors · doi:10.3390/s23031388 · Cited by 7
“There is another type of test, called a synchronized test, which aims to study collective capabilities of smartphones in executing a compute-intensive scenario, for example, a set of distributable workloads from a specialized AI application such as object recognition from images using neural networks [9].”
BackgroundSciNet: Codesign of Resource Management in Cloud Computing Environments
Tuli, S., Casale, G., & Jennings, N. R. (2023) · IEEE Transactions on Computers · doi:10.1109/tc.2023.3310678 · Cited by 10
“Deployment deals with choosing the optimal AI model or distribution strategy for optimal accuracy and response times [7].”
BackgroundGQFedWAvg: Optimization-Based Quantized Federated Learning in General Edge Computing Systems
Li, Y., Cui, Y., & Lau, V. K. N. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3335302 · Cited by 7
“R ECENT years have witnessed a growing interest in federated learning (FL) in general edge computing systems, where the computing and communication resources at the workers may be uniform or distinct [2], [3], [4].”
BackgroundOptimizing Federated Learning with Heterogeneous Edge Devices
Islam, M. M., & Alawad, M. (2024) · IEEE International Conference on Computing and Machine Intelligence (ICMI) · doi:10.1109/icmi60790.2024.10585747 · Cited by 5
“However, training DL models for IoT necessitates a sizable dataset of labeled examples that encompass diverse scenarios encountered in real-world IoT environments [2].”
BackgroundChen, X., Wang, A., Deng, X., & Gui, J. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3462722 · Cited by 6
“These challenges can be mitigated by harnessing the computational resources available on edge devices, enabling training directly at the network’s edge without the need to share data [1].”
Edge-device collaborative computing for multi-view classification
Palena, M., Cerquitelli, T., & Chiasserini, C. F. (2024) · Computer Networks · doi:10.1016/j.comnet.2024.110823 · Cited by 7
“Differently from existing works tackling the orchestration of DNNs at the edge [1, 2] or the offloading of such tasks from end devices to the edge [2, 3], our first goal is to leverage the redundancy or overlap of the images collected by such cameras to increase the accuracy of the inference while…”
BackgroundDeep Learning Inference on Edge: A Preliminary Device Comparison
González, M. L., Ruiz, J., Andrés, L., Lozada, R., Skibinsky-Gitlin, E. S., Fernández, J. O., Sedano, J., & García-Vico, Á. M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-77731-8_25 · Cited by 6
A Computer Vision-Based Dual Network Approach for Indoor Fall Detection
Zhang, L., Fang, C., & Zhu, M. J. (2020) · International Journal of Innovative Science and Research Technology (IJISRT) · doi:10.38124/ijisrt20aug551 · Cited by 5
Geolocation-Centric Information Platform for Resilient Spatio-temporal Content Management
Tsukamoto, K., Tamura, H., Taenaka, Y., Nobayashi, D., Yamamoto, H., Ikenaga, T., & Lee, M. J. (2020) · IEICE Transactions on Communications · doi:10.1587/transcom.2020nvi0003 · Cited by 10
Cloud-backed mobile cognition: Power-efficient deep learning in the autonomous vehicle era
Vega, A. J., Buyuktosunoglu, A., Callegaro, D., Levorato, M., & Bose, P. K. (2021) · Computing · doi:10.1007/s00607-021-00953-7 · Cited by 5
“The deployment of DL applications in these scenarios garnered significant attention with focus on on-device and cloud-backed deep learning computation [3], and consideration of bandwidth optimization aspects.”
BackgroundEdge-Cloud Architectures Using UAVs Dedicated To Industrial IoT Monitoring And Control Applications
Salhaoui, M., González, A. G., Arioua, M., Molina, J. C. M., Ortiz, F. J., & Oualkadi, A. E. (2020) · International Symposium on Advanced Electrical and Communication Technologies (ISAECT) · doi:10.1109/isaect50560.2020.9523700 · Cited by 7
Optimization of MLP Neural Networks in 8-bit Microcontrollers using Program Memory
Guimarães, C. J. B. V., Torquato, M. F., & Fernandes, M. A. C. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9533594 · Cited by 5
“The authors from [13]–[16] discussed the importance of Edge Computing (also known as Fog Computing) applications.”
BackgroundAnalyse or Transmit: Utilising Correlation at the Edge with Deep Reinforcement Learning
Hribar, J., Shinkuma, R., Iosifidis, G., & Dusparić, I. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685166 · Cited by 5
Adaptive In-Network Collaborative Caching for Enhanced Ensemble Deep Learning at Edge
Qin, Y., Wu, D., Xu, Z., Tian, J., & Zhang, Y. (2021) · Mathematical Problems in Engineering · doi:10.1155/2021/9285802 · Cited by 9
DePo: Dynamically Offload Expensive Event Processing to the Edge of Cyber-Physical Systems
Ma, M., Zhang, J., & Wang, P. (2021) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2021.3135441 · Cited by 5
“The related research of DNN partitioning [36] provides some clues to this problem.”
BackgroundUbi-Flex-Cloud: ubiquitous flexible cloud computing: status quo and research imperatives
Thyagaturu, A. S., Nguyen, G. T. K., Rimal, B. P., & Reisslein, M. (2022) · Applied Computing and Informatics · doi:10.1108/aci-02-2022-0029 · Cited by 8
Chen, S., Mathew, J. J., Feng, C.-T., & Hsu, T.-J. (2022) · International Conference on Applied System Innovation · doi:10.1109/icasi55125.2022.9774445 · Cited by 5
Miryala, S., Zaman, M. A., Mittal, S. S., Ren, Y., Deptuch, G. W., Carini, G. A., Zohar, S., Yoo, S., Fried, J., Huang, J., et al. (2022) · International Symposium on Quality Electronic Design (ISQED) · doi:10.1109/isqed54688.2022.9806285 · Cited by 5
A System Deployment Model of Multi-CCD Automatic Optical Inspection for Economical Operations
Wang, K., & Qiu, Y.-C. (2022) · IEEE Access · doi:10.1109/access.2022.3178999 · Cited by 6
Migrating Deep Learning Data and Applications among Kubernetes Edge Nodes
Mangkhangcharoen, S., Haga, J., & Rattanatamrong, P. (2021) · IEEE International Conference on High Performance Computing and Communications; IEEE International Conference on Smart City; IEEE International Conference on Data Science and Systems (HPCC/SmartCity/DSS) · doi:10.1109/hpcc-dss-smartcity-dependsys53884.2021.00299 · Cited by 8
AI Workload Allocation Methods for Edge-Cloud Computing: A Review
Rafea, S. A., & Jasim, A. D. (2023) · Al-Iraqia Journal of Scientific Engineering Research · doi:10.58564/ijser.2.4.2023.125 · Cited by 6
A Smartphone-Based Cell Segmentation to Support Nasal Cytology
Dimauro, G., Pierro, D. D., Deperte, F., Simone, L., & Fina, P. R. (2020) · Applied Sciences · doi:10.3390/app10134567 · Cited by 5
Bayesian Online Learning for MEC Object Recognition Systems
Galanopoulos, A., Ayala‐Romero, J. A., Iosifidis, G., & Leith, D. J. (2020) · Globecom · doi:10.1109/globecom42002.2020.9322146 · Cited by 7
“While numerous studies have investigated optimizing the performance of edge computing systems [3], the majority focus on throughput, delay, and other general criteria.”
BackgroundEmbedding Sequence Model in STM32 Based Neuro-Controller
Sinkevych, O., Boyko, Y. V., Rechynskyi, O., Sokolovskiĭ, B. S., & Monastyrskii, L. S. (2021) · IEEE International Conference on Electronics and Information Technologies (ELIT) · doi:10.1109/elit53502.2021.9501132 · Cited by 7
SplitPlace: Intelligent Placement of Split Neural Nets in Mobile Edge Environments
Tuli, S. (2022) · ACM SIGMETRICS Performance Evaluation Review · doi:10.1145/3512798.3512821 · Cited by 8
“This is seldom the case for low-end nodes like Raspberry Pis [14].”
BackgroundAdvances in Multi‐modal Intelligent Sensing
Rehman, M. U., Jamshed, M. A., & Kalsoom, T. (2024) · Wiley eBooks · doi:10.1002/9781394257744.ch1 · Cited by 14
Towards lightweight deep neural network for smart agriculture on embedded systems
Du, P., Polonelli, T., Magno, M., & Cheng, Z. (2022) · 2022 IEEE Sensors Applications Symposium (SAS) · doi:10.1109/sas54819.2022.9881382 · Cited by 5
“Thus, these algorithms are suitable for running on servers or embedded platforms with high power consumption in the range of Watts, and cannot be deployed to low-cost, low-power, and battery-supplied sensors remotely connected to a server [7].”
Automatic pose estimation in newborn infants: Lessons from the Baby Grow study
Sotoodeh, M. S., Ossmy, O., Donati, G., Hall, J., Rowan, H., & Forrester, G. S. (2026) · Behavior Research Methods · doi:10.3758/s13428-026-02943-z · Cited by 4
“RTMpose is also compatible with RTMdet for drawing bounding boxes; however, we found that it is less stable and makes redundant detections in frames in comparison to MMDetection (please see supplementary file S6 for further information and examples).”
InfluentialBibi, M., Khan, Q. W., Yazdan, S. A., Ahmad, R., & Kim, D.-H. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.104798 · Cited by 4
Cai, F., Yuan, D., Xie, M., He, W. P., Kong, L., Guo, W., Jiang, Y., & Cui, L. (2023) · IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws60048.2023.00045 · Cited by 7
Communication Technologies and Security Challenges in IoT: An Introduction
Kumar, A., Prasad, A., & Singh, T. P. (2024) · Internet of things · doi:10.1007/978-981-97-0052-3_1 · Cited by 5
Improving YOLO Object Detection Performance on Single-Board Computer using Virtual Machine
Haq, M. A., Huy, L. N. Q., & Fahriani, N. (2024) · Emerging Information Science and Technology · doi:10.18196/eist.v5i1.22486 · Cited by 7
“Deep learning has become a critical component in the evolution of computing technology, notably impacting the functionality and efficiency of single-board computers (SBCs) [1].”
BackgroundLi, X., & Chen, M. (2024) · PeerJ Computer Science · doi:10.7717/peerj-cs.2306 · Cited by 5
Efficient quantized transformer for atrial fibrillation detection in cross-domain datasets
Toosi, M. H., Mohammadi-nasab, M., Mohammadi, S., & Salehi, M. E. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110371 · Cited by 6
Yang, M. K., & Jeong, J. (2025) · Applied Sciences · doi:10.3390/app15073863 · Cited by 9
“As data size and complexity grow, efficient computational methods and strong encryption algorithms have become indispensable [5,6].”
Asynchronous Distributed Gaussian Process Regression
Yang, Z., Dai, X., & Hirche, S. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i21.34359 · Cited by 10
Unsupervised Federated Learning for Face Recognition in Decentralized Environments
Solomon, E. (2025) · International Conference on Information and Computer Technologies (ICICT) · doi:10.1109/icict64582.2025.00022 · Cited by 4
“Mostly, these systems leverage ML methodologies to train deep neural networks using facial data sets collected primarily from end devices such as smartphones, with centralized model training on servers [1], [33], [34].”
LUCID: A Framework for Reducing False Positives and Inconsistencies Among Container Scanning Tools
Haq, M. S., Tosun, A. Ş., & Korkmaz, T. (2025) · IEEE Access · doi:10.1109/access.2025.3587626 · Cited by 10
“The application of multiple filters allows us to see how the inconsistencies depend on certain criteria and on each other.”
Edge computing for IoT: Novel insights from a comparative analysis of access control models
Xue, T., Zhang, Y., Wang, Y., Wang, W., Li, S., & Zhang, H. (2025) · Computer Networks · doi:10.1016/j.comnet.2025.111468 · Cited by 15
An Intelligent Doorbell Design Using Federated Deep Learning
Patel, V., Kanani, S., Pathak, T., Patel, P., Ali, M. I., & Breslin, J. G. (2020) · Joint International Conference on Data Science & Management of Data (CODS-COMAD) · doi:10.1145/3430984.3430988 · Cited by 4
“This approach employs various model reduction techniques [2] (e.”
MethodDistributed training for accelerating metalearning algorithms
Kunde, S., Pandit, A., Mishra, M., & Singhal, R. (2021) · International Workshop on Big Data in Emergent Distributed Environments (BiDEDE) · doi:10.1145/3460866.3461773 · Cited by 4
Rodríguez, A., Otero, A., Platzner, M., & de la Torre, E. (2021) · IEEE Transactions on Computers · doi:10.1109/tc.2021.3107196 · Cited by 6
Side-Channel Analysis-Based Model Extraction on Intelligent CPS: An Information Theory Perspective
Pan, Q., Wu, J., Lin, X., & Li, J. (2021) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings-greencom-cpscom-smartdata-cybermatics53846.2021.00050 · Cited by 5
Randellini, E., Rigutini, L., & Saccà, C. (2022) · International Journal of Artificial Intelligence & Applications · doi:10.5121/ijaia.2022.13104 · Cited by 7
“Edge computing is a viable solution to meet the latency, scalability and privacy challenges described above [41].”
BackgroundDeep Learning on FPGAs with Multiple Service Levels for Edge Computing
Gao, C., Saha, S., Lu, Y., Saha, R., McDonald-Maier, K. D., & Zhai, X. (2022) · International Conference on Automation and Computing (ICAC) · doi:10.1109/icac55051.2022.9911081 · Cited by 4
“Deep Neural Network accelerators based on field-programmable gate arrays (FPGA) are another promising method, as FPGA can enable fast computing while keeping reconfigurability [11].”
BackgroundMethodAn Energy-Efficient Method for Recurrent Neural Network Inference in Edge Cloud Computing
Chen, C., Guo, W., Wang, Z., Yang, Y., Wu, Z., & Li, G. (2022) · Symmetry · doi:10.3390/sym14122524 · Cited by 7
Let’s Unleash the Network Judgment: A Self-Supervised Approach for Cloud Image Analysis
Demattíes, D., Raut, B., Park, S., Jackson, R. C., Shahkarami, S., Kim, Y., Sankaran, R., Beckman, P., Collis, S., & Ferrier, N. (2023) · Artificial Intelligence for the Earth Systems · doi:10.1175/aies-d-22-0063.1 · Cited by 5
“Edge computing (Wang et al. 2020b; Chen and Ran 2019; Beckman et al. 2016; Cao et al. 2020) is a new processing paradigm that collocates powerful and efficient computing hardware and the sensors that provide the data.”
BackgroundReducing environment exposure to COVID-19 by IoT sensing and computing with deep learning
Ma, C., Song, J., Xu, Y., Fan, H., Liu, X., Wu, X., Luo, Y., Sun, T., & Xie, J. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08712-9 · Cited by 4
“Edge devices provide the basis to the real-time collect large amount of data and achieve analysis with deep learning [6].”
BackgroundToma, A., Samara, S., & Qamhieh, M. (2023) · Journal of Network and Systems Management · doi:10.1007/s10922-023-09770-z · Cited by 6
“The models in [15, 22, 23] considered solutions from machine learning which may have high overhead, especially in dynamic environments with uncertainty or fluctuating parameters, and might dominate performance enhancement consequently.”
BackgroundMethodFalling Behavior Detection System for Elevator Passengers Based on Deep Learning and Edge Computing
Chen, Y., Zhao, Q., Fan, Q., Huang, X., Wu, F., & Qi, J. (2023) · Journal of Physics Conference Series · doi:10.1088/1742-6596/2644/1/012012 · Cited by 4
“As shown in Figure 1, edge computing places computing nodes in a fine-grained mesh closer to the end devices, making it a feasible approach to meet the requirements of high computation and low latency for edge devices in deep learning [8].”
BackgroundHu, Z., Zhao, X., Guo, Y., & Zhou, C. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3347616 · Cited by 3
“In addition, the model parameter of TFDR-PNN is 11 668, which is smaller than most of the edge models reviewed in the literature [34].”
ResultKim, C., Kim, G.-Y., & Kim, S. (2023) · International Conference on Information and Communication Technology Convergence (ICTC) · doi:10.1109/ictc58733.2023.10392296 · Cited by 4
“Specifically, achieving high levels of accuracy through AI techniques in domains like object detection and anomaly detection requires substantial computational resources and memory during both the training and inference phases[1].”
BackgroundZhang, X., Hu, Z., Liang, Y., Xiao, H., Xu, A., Zheng, M., & Sun, C. (2024) · Journal of Grid Computing · doi:10.1007/s10723-023-09730-6 · Cited by 4
“However, these methods often overlook the issues of inter-model collaboration and data privacy protection [42].”
Investigation on big data evaluation and visualization of internet of things based on edge computing
Long, Q. (2024) · Measurement Sensors · doi:10.1016/j.measen.2024.101177 · Cited by 8
A deep learning based approach for image retrieval extraction in mobile edge computing
Alasadi, J., Bati, G. F., & Hilli, A. A. (2024) · Journal of Umm Al-Qura University for Engineering and Architecture · doi:10.1007/s43995-024-00060-6 · Cited by 4
“The training phase is computationally expensive due to optimization of millions of parameters that must be repeated many times [3].”
Advances in Few-Shot Action Recognition: A Comprehensive Review
Ruan, Z., Wei, Y., Yuan, Y., Li, Y., Guo, Y., & Xie, Y. (2024) · International Conference on Artificial Intelligence and Big Data (ICAIBD) · doi:10.1109/icaibd62003.2024.10604585 · Cited by 4
“Moreover, traditional deep learning algorithms may perform well on a specific task, but even if the datasets are similar, the trained models cannot be easily migrated to other tasks [4].”
Background
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.