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 201–300 of 1,000
Maintaining Privacy in Face Recognition Using Federated Learning Method
Woubie, A., Solomon, E., & Attieh, J. (2024) · IEEE Access · doi:10.1109/access.2024.3373691 · Cited by 47
Wearable hydrogel-based health monitoring systems: A new paradigm for health monitoring?
Wang, X., Ji, H., Gao, L., Hao, R., Shi, Y., Yang, J., Hao, Y., & Chen, J. (2024) · Chemical Engineering Journal · doi:10.1016/j.cej.2024.153382 · Cited by 60
Self-citationNanophotonic structure inverse design for switching application using deep learning
Adibnia, E., Ghadrdan, M., & Mansouri-Birjandi, M. A. (2024) · Scientific Reports · doi:10.1038/s41598-024-72125-4 · Cited by 43
“Additional considerations, including overfitting 39 , issues of model interpretability 40,41 , computational requisites 42,43 , susceptibility to adversarial attacks 44,45 , and obstacles in inverse design 46 characterized by the many-to-one dilemma 47,48 , underscore the imperative for continual…”
BackgroundA Latency-Optimized Reconfigurable NoC for In-Memory Acceleration of DNNs
Mandal, S. K., Krishnan, G., V, Chakrabarti, C., Seo, J.-S., Cao, Y. K., & Ogras, U. Y. (2020) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2020.3015509 · Cited by 40
“For example, authors in [17] show the extensive usage of LeNet, SqueezeNet, VGG in various edge devices, and [18] use ResNet-152 DNN for cloud-based applications such as video analytics.”
Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
Prashanthi, S. K., Kesanapalli, S. A., & Simmhan, Y. L. (2022) · Proceedings of the ACM on Measurement and Analysis of Computing Systems · doi:10.1145/3570604 · Cited by 66
“(2) We study the impact of storage medium and mini-batch sizes on stalls, GPU compute time and end-to-end times (Sec.”
Multi-Cell Multi-Beam Prediction Using Auto-Encoder LSTM for mmWave Systems
Shah, S. H. A., & Rangan, S. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3183632 · Cited by 50
“Similar architecture is proposed in [12].”
IoT-based edge computing (IoTEC) for improved environmental monitoring
Roostaei, J., Wager, Y. Z., Shi, W., Dittrich, T. M., Miller, C. J., & Gopalakrishnan, K. K. (2023) · Sustainable Computing Informatics and Systems · doi:10.1016/j.suscom.2023.100870 · Cited by 73
Edge Intelligence for Smart Grid: A Survey on Application Potentials
Gooi, H. B., Wang, T., & Tang, Y. (2023) · CSEE Journal of Power and Energy Systems · doi:10.17775/cseejpes.2022.02210 · Cited by 78
“EC is a distributed computing paradigm that brings computation and data storage closer to data sources, which can avoid unnecessary communication latency and enable faster responses for end-users [14], [15].”
Big Data Analytics Framework for Decision-Making in Sports Performance Optimization
Mănescu, D. C. (2025) · Data · doi:10.3390/data10070116 · Cited by 76
“…Edge deploy lightweight models (TensorFlow Lite, Open Neural Network Exchange (ONNX) Runtime) for on-device inference (e.g., PlayerLoad™, fatigue detection), reducing data volume by up to 80% and, under optimized conditions, enabling sub-100 ms decision cycles through localized edge inference [13].”
Mobile app recommendation via heterogeneous graph neural network in edge computing
Liang, T., Sheng, X., Zhou, L., Li, Y., Gao, H., Yin, Y., & Chen, L. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.107162 · Cited by 41
Intelligent Service Migration Based on Hidden State Inference for Mobile Edge Computing
Li, X., Chen, S., Zhou, Y., Chen, J., & Feng, G. (2021) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2021.3103511 · Cited by 44
Self-citationSemi-supervised Federated Learning for Activity RecognitionPreprint
Zhao, Y., Liu, H., Li, H., Barnaghi, P. M., & Haddadi, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.00851 · Cited by 76
“Modern edge devices such as Intel Next Unit of Computing (NUC) [1] and Raspberry Pi [2] are capable of running DNN models [8, 29] and providing real-time activity recognition [6, 21] from videos.”
BackgroundZhou, X., Gao, Y., Li, C., & Huang, Z. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3067454 · Cited by 46
“…manner and it is necessary to offload multiple objective tasks to multiple edge servers for optimizing the edge device’s memory, computation and communication demands, which can reduce response time, provide more efficient processing, and alleviate the pressure on the network [22], [23], [24].”
BackgroundJanbi, N. F., Mehmood, R., Katib, I. A., Albeshri, A. A., Corchado, J. M., & Yiğitcanlar, T. (2022) · Sensors · doi:10.3390/s22051854 · Cited by 46
Zhang, W., Zeadally, S., Li, W., Zhang, H., Hou, J., & Leung, V. C. M. (2022) · IEEE Transactions on Cloud Computing · doi:10.1109/tcc.2022.3175725 · Cited by 52
“Although resource allocation and data management have been widely used for improving latency and energy effi-ciency for edge AI [8], [11], how to efficiently allocate the resources by considering both the impact of data quality on computation complexity and result quality has not been studied in…”
BackgroundMethodInfluentialReview of diffractive deep neural networks
Sun, Y., Dong, M., Yu, M., Liu, X., & Zhu, L. (2023) · Journal of the Optical Society of America B · doi:10.1364/josab.497148 · Cited by 50
Quttainah, M. A., Mishra, V., Madakam, S., Lurie, Y., & Mark, S. (2024) · JMIR AI · doi:10.2196/51834 · Cited by 61
“Therefore, the cost of training an LLM for medical education may be prohibitive [19].”
BackgroundA Comprehensive Survey on Federated Learning in the Healthcare Area: Concept and Applications
Upreti, D., Yang, E., Kim, H., & Seo, C. (2024) · Computer Modeling in Engineering & Sciences · doi:10.32604/cmes.2024.048932 · Cited by 59
“An overview of federated learning systems 1) Client devices: These are the edge devices that hold the data used for training the model [24].”
Federated learning for anomaly detection on Internet of Medical Things: A survey
Pinto, R. P., Silva, B. M. C., & Inácio, P. R. M. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101677 · Cited by 52
Achieving Democracy in Edge Intelligence: A Fog-Based Collaborative Learning Scheme
Zhang, T., Shen, Z., Jin, J., Zheng, X., Tagami, A., & Cao, X. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3020911 · Cited by 41
“…focus on using the cloud-fog stereo-type to complete the training (either training the model at the cloud or using the cloud as the parameter server [5]), there is little attention on confining the model training and inference process entirely at the fog plane without the reliance on the cloud.”
MethodEdge Learning: The Enabling Technology for Distributed Big Data Analytics in the Edge
Zhang, J., Qu, Z., Chen, C., Wang, H., Zhan, Y., Ye, B., & Guo, S. (2021) · ACM Computing Surveys · doi:10.1145/3464419 · Cited by 83
“In [158] and [13], Zhou et al. provide an overview of the architectures, frameworks, and emerging key technologies for deep learning model toward training/inference at the network edge.”
BackgroundAutonomous Driving with Deep Learning: A Survey of State-of-Art TechnologiesPreprint
Huang, Y., & Chen, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.06091 · Cited by 115
“It is seen existing cloud computing is unable to manage distributed learning tasks, edge computing [34, 35] turns out to be an alternative.”
BackgroundLema, D. G., Usamentiaga, R., & García, D. F. (2024) · Integration · doi:10.1016/j.vlsi.2023.102127 · Cited by 41
Brain-inspired computing systems: a systematic literature review
Zolfagharinejad, M., Alegre-Ibarra, U., Chen, T., Kinge, S., & van der Wiel, W. G. (2024) · The European Physical Journal B · doi:10.1140/epjb/s10051-024-00703-6 · Cited by 39
Local Scheduling in KubeEdge-Based Edge Computing Environment
Kim, S.-H., & Kim, T. (2023) · Sensors · doi:10.3390/s23031522 · Cited by 45
“EdgeGossip is implemented on the Gossip algorithm [22], and its effectiveness was demonstrated using real-time deep-learning workloads.”
MethodA Review of On-Device Machine Learning for Iot: An Energy
Tekin, N., Arış, A., Acar, A., Uluagac, A. S., & Güngör, V. Ç. (2023) · Ad Hoc Networks · doi:10.1016/j.adhoc.2023.103348 · Cited by 50
Martinez, A. D., Ser, J. D., Villar-Rodríguez, E., Osaba, E., Poyatos, J., Tabik, S., Molina, D., & Herrera, F. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.10.014 · Cited by 45
Generative Adversarial Super-Resolution at the edge with knowledge distillation
Angarano, S., Salvetti, F., Martini, M., & Chiaberge, M. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106407 · Cited by 46
Ni, X., Takeda, F., Jiang, H., Yang, W. Q., Saito, S., & Li, C. (2022) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2022.107200 · Cited by 40
Analysis of Machine Learning Algorithms for Anomaly Detection on Edge Devices
Huč, A., Šalej, J., & Trebar, M. (2021) · Sensors · doi:10.3390/s21144946 · Cited by 55
“Deep learning is widely used and has been very successful across various application domains [4].”
Enabling DNN Acceleration With Data and Model Parallelization Over Ubiquitous End Devices
Huang, Y., Qiao, X., Lai, W., Dustdar, S., Zhang, J., & Li, J. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3112715 · Cited by 39
“D EEP learning [e.g., deep neural networks (DNNs)] is currently a representative way of achieving artificial intelligence (AI) in numerous applications [1]–[3].”
BackgroundSurvey on Mobile Edge-Cloud Computing: A Taxonomy on Computation offloading Approaches
Elgendy, I. A., & Yadav, R. (2021) · Studies in big data · doi:10.1007/978-3-030-85428-7_6 · Cited by 35
Luo, H., Chen, T., Li, X., Li, S., Zhang, C., Zhao, G., & Liu, X. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3157957 · Cited by 43
“Besides, knowledge distillation has become an important branch in the field of DNN compression and acceleration [13], [21].”
BackgroundEstimating crowd density with edge intelligence based on lightweight convolutional neural networks
Wang, S., Pu, Z., Li, Q., & Wang, Y. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117823 · Cited by 46
Joint multi-user DNN partitioning and task offloading in mobile edge computing
Liao, Z., Hu, W., Huang, J., & Wang, J. (2023) · Ad Hoc Networks · doi:10.1016/j.adhoc.2023.103156 · Cited by 42
Artificial intelligence assisted nanogenerator applications
Xu, S., Manshaii, F., Xiao, X., & Chen, J. (2024) · Journal of Materials Chemistry A · doi:10.1039/d4ta07127a · Cited by 38
Self-citationServing at the Edge: An Edge Computing Service Architecture Based on ICN
Fan, Z., Yang, W., Wu, F., Cao, J. X., & Shi, W. (2021) · ACM Transactions on Internet Technology · doi:10.1145/3464428 · Cited by 39
“With the adventure of edge computing, to improve the real-time responsiveness of AI services, machine learning models are usually deployed to the edge of the network to implement edge intelligence [8, 46].”
BackgroundClark, M. L., Salas, L. A., Baligar, S., Quinn, C. A., Snyder, R. L., Leland, D., Schackwitz, W., Goetz, S. J., & Newsam, S. D. (2023) · Ecological Informatics · doi:10.1016/j.ecoinf.2023.102065 · Cited by 52
“There is growing interest in applying CNN classifiers for bird vocalization classification/detection in situations at the edge of a data computation network, such as directly on a smartphone, ARU, or Internet of Things device (Chen and Ran, 2019).”
BackgroundFedKNOW: Federated Continual Learning with Signature Task Knowledge Integration at Edge
Luopan, Y., Han, R., Zhang, Q., Liu, C. H., Wang, G., & Chen, L. Y. (2023) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde55515.2023.00033 · Cited by 46
“DNN models to retrain and adapt to those changes [6].”
Consumers profiling based federated learning approach for energy load forecasting
Dogra, A., Anand, A., & Bedi, J. (2023) · Sustainable Cities and Society · doi:10.1016/j.scs.2023.104815 · Cited by 34
Pääkkönen, P., & Pakkala, D. (2020) · Journal Of Big Data · doi:10.1186/s40537-020-00303-y · Cited by 44
“Often, training performance has been improved by optimising the frequency and size of training updates [8].”
BackgroundMethodInfluentialEfficiency Versus Accuracy: A Review of Design Techniques for DNN Hardware Accelerators
Latotzke, C., & Gemmeke, T. (2021) · IEEE Access · doi:10.1109/access.2021.3050670 · Cited by 39
“In many fields, the analytical information extraction requires high security and low latency [1], [2].”
BackgroundTEA-fed: time-efficient asynchronous federated learning for edge computing
Zhou, C., Tian, H., Zhang, H., Zhang, J., Dong, M., & Jia, J. (2021) · ACM International Conference on Computing Frontiers (CF) · doi:10.1145/3457388.3458655 · Cited by 61
“In the era of big data, edge computing is an effective solution to the communication delay, data scalability, and privacy challenges brought by traditional cloud computing[1].”
BackgroundDynamic and adaptive fault-tolerant asynchronous federated learning using volunteer edge devices
Morell, J. Á., & Alba, E. (2022) · Future Generation Computer Systems · doi:10.1016/j.future.2022.02.024 · Cited by 41
E-prop on SpiNNaker 2: Exploring online learning in spiking RNNs on neuromorphic hardware
Rostami, A., Vogginger, B., Yan, Y., & Mayr, C. (2022) · Frontiers in Neuroscience · doi:10.3389/fnins.2022.1018006 · Cited by 53
“By omitting the communication between edge and cloud, we can reduce power consumption of devices and increase their lifetimes (Sze et al., 2017; Chen and Ran, 2019; Cai et al., 2020).”
BackgroundDistributed Intelligence in Wireless Networks
Liu, X., Yu, J., Liu, Y., Gao, Y., Mahmoodi, T., Lambotharan, S., & Tsang, D. H. K. (2023) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2023.3265425 · Cited by 48
“Particularly, The authors in [31] focused on an overview of the deep learning applications at the network edge.”
BackgroundKim, J., Kim, G., Yoshitoshi, R., & Tokuda, K. (2025) · Sensors · doi:10.3390/s25154586 · Cited by 47
“Including them in our study would necessarily shift our research focus from architectural paradigm analysis to optimization benchmarking [30], requiring fundamentally different methodologies and evaluation criteria.”
Remaining useful life prediction based on state assessment using edge computing on deep learning
Hsu, H.-Y., Srivastava, G., Wu, H., & Chen, M. (2020) · Computer Communications · doi:10.1016/j.comcom.2020.05.035 · Cited by 37
Stable and compact design of Memristive GoogLeNet Neural Network
Ran, H., Wen, S., Shi, K., & Huang, T. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.01.122 · Cited by 36
Cross-Camera Inference on the Constrained Edge
Li, J., Liu, L., Xu, H. L., Wu, S., & Xue, C. J. (2023) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom53939.2023.10229045 · Cited by 41
“Utilizing energy efficiently is critical to the applications run on the constrained edge nodes [13].”
Stochastic Cumulative DNN Inference With RL-Aided Adaptive IoT Device-Edge Collaboration
Qu, K., Zhuang, W., Wu, W. R., Li, M., Shen, X., Li, X., & Shi, W. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3280746 · Cited by 38
“Recently, collaborative inference over a device–edge–cloud computing hierarchy has been investigated to improve the delay performance, by partially offloading the DNN model computation to nearby devices, edge server, or cloud [8], [9], [10], [11].”
BackgroundCollaborative Edge Intelligence for Autonomous Vehicles: Opportunities and Challenges
Bai, L., Cao, J., Zhang, M., & Li, B. (2025) · IEEE Network · doi:10.1109/mnet.2024.3520166 · Cited by 34
“According to Cisco, 75% of these data will be generated and processed at the network edge by 2025 [14].”
A Generalization Performance Study Using Deep Learning Networks in Embedded Systems
Gorospe, J., Mulero, R., Arbelaitz, O., Muguerza, J., & Antón, M. Á. (2021) · Sensors · doi:10.3390/s21041031 · Cited by 32
“Once the significance of both EC and DL has been contrasted separately, the next step is to check the current state of the combination of the two areas [45], focusing the analysis on the on-device intelligence architecture.”
BackgroundJoint Coding and Scheduling Optimization for Distributed Learning Over Wireless Edge Networks
Huynh, N. V., Hoang, D. T., Nguyen, D. N., & Dutkiewicz, E. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118432 · Cited by 34
edge–SR: Super–Resolution For The Masses
Michelini, P. N., Lu, Y., & Jiang, X. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00407 · Cited by 37
ENTS: An Edge-native Task Scheduling System for Collaborative Edge Computing
Zhang, M., Cao, J., Yang, L., Zhang, L., Sahni, Y., & Jiang, S. (2022) · IEEE/ACM Symposium on Edge Computing (SEC) · doi:10.1109/sec54971.2022.00019 · Cited by 43
“Compared with traditional cloud computing, the emerging edge computing paradigm enjoys outstanding benefits, including reduced response latency and enhanced privacy preservation [2] [3].”
BackgroundReggiani, E., Pappalardo, A., Doblas, M., Moretó, M., Olivieri, M., Ünsal, O., & Cristal, A. (2023) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca56546.2023.10071076 · Cited by 54
“Accelerating quantized DNNs represents a widespread research topic [16], [20], [30].”
BackgroundOn Edge Computing for Remote Pathology Consultations and Computations
Sacco, A., Esposito, F., Marchetto, G., Kolar, G. R., & Schwetye, K. E. (2020) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.3007661 · Cited by 32
“By low-latency system we mean a program able to reduce latency to the order of hundreds of milliseconds, considered as a threshold for enabling the user interaction [26].”
BackgroundIncreasing Traffic Safety with Real-Time Edge Analytics and 5G
Lujic, I., Maio, V. D., Pollhammer, K., Bodrozic, I., Lasic, J., & Brandić, I. (2021) · International Workshop on Edge Systems, Analytics and Networking (EdgeSys) · doi:10.1145/3434770.3459732 · Cited by 42
“Compared to traditional scenarios of sending sensitive data over the network to the cloud, we should ensure that no sensitive data are stored or sent over the network [4].”
BackgroundA Benchmark of Deep Learning Models for Multi-leaf Diseases for Edge Devices
Anh, P. T., & Duc, H. T. M. (2021) · International Conference on Advanced Technologies for Communications · doi:10.1109/atc52653.2021.9598196 · Cited by 34
Meyer, J., Frank, A., Schlebusch, T., & Kasneci, E. (2021) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3494998 · Cited by 34
“To reduce them a further network optimization could be achieved by model pruning or other optimization techniques [12].”
MethodA Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network InferencePreprint
Hong, S., Kaya, Y., Modoranu, I.-V., & Dumitraş, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.02432 · Cited by 77
“However, complex deep learning models are impractical for low-power edge devices, such as an Arduino, that are common in the IoT scenarios (Chen & Ran, 2019).”
BackgroundHuman-centred artificial intelligence for mobile health sensing: challenges and opportunities
Dang, T., Spathis, D., Ghosh, A., & Mascolo, C. (2023) · Royal Society Open Science · doi:10.1098/rsos.230806 · Cited by 30
“The breakthroughs in efficient deep learning training revolve around optimizing training update frequencies between local and server devices, determining which part of the model needs an update, and so forth [199].”
BackgroundDISNET: Distributed Micro-Split Deep Learning in Heterogeneous Dynamic IoT
Samikwa, E., Maio, A. D., & Braun, T. I. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3313514 · Cited by 36
“Deep learning (DL) driven applications are computation-intensive because of the depth of the DNN models (i.e., their massive number of layers) and large input dimen-sion [9], [17], [18].”
BackgroundCRIME: Input-Dependent Collaborative Inference for Recurrent Neural Networks
Pagliari, D. J., Chiaro, R., Macii, E., & Poncino, M. (2020) · IEEE Transactions on Computers · doi:10.1109/tc.2020.3021199 · Cited by 36
“However, pure cloudcomputing raises several issues [10].”
BackgroundCOTAF: Convergent Over-the-Air Federated Learning
Sery, T., Shlezinger, N., Cohen, K., & Eldar, Y. C. (2020) · Globecom · doi:10.1109/globecom42002.2020.9322580 · Cited by 34
“Since data samples are available at remote users, e.g., smartphones and other edge devices, the common strategy is to gather these samples at a computationally powerful server, which uses them to train its model [2].”
BackgroundLightweight Compression of Intermediate Neural Network Features for Collaborative Intelligence
Cohen, R. A., Choi, H., & Bajić, I. V. (2021) · IEEE Open Journal of Circuits and Systems · doi:10.1109/ojcas.2021.3072884 · Cited by 37
“For devices located at the edge of a network, lightweight and mobile-friendly architectures [1]–[3] can facilitate the implementation of these DNNs.”
BackgroundDEFER: Distributed Edge Inference for Deep Neural Networks
Parthasarathy, A., & Krishnamachari, B. (2022) · International Conference on Communication Systems & Networks (COMSNETS) · doi:10.1109/comsnets53615.2022.9668515 · Cited by 37
“Therefore, researchers and practitioners have advocated performing inference at the edge, close to the IoT data sources [1].”
Edge-Computing-Based Knowledge Distillation and Multitask Learning for Partial Discharge Recognition
Ji, J., Shu, Z., Li, H., Lai, K. X., Lu, M., Jiang, G., Wang, W., Zheng, Y., & Jiang, X. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3351239 · Cited by 29
“These algorithms demand large volumes of data to learn the input-output mappings required for accurate predictions and make training a slow and resource-intensive process [18], [19], [20], [21].”
BackgroundFlexible nanogenerators for intelligent robotics: design, manufacturing, and applications
Zhao, H., Ruan, L., Wang, Z., Shu, M., Lyu, C., Bulathsinghala, R. L., Dharmasena, I., Wu, C., & Ding, W. (2024) · International Journal of Extreme Manufacturing · doi:10.1088/2631-7990/ad94b8 · Cited by 41
“Furthermore, implementing edge computing solutions for local data processing will further enhance the functionality and autonomy of these sensors by allowing them to process data on-site, reducing latency and dependence on centralized computing resources [244].”
Distributed machine learning for energy trading in electric distribution system of the future
Wang, N., Li, J., Ho, S.-S., & Qiu, C. (2020) · The Electricity Journal · doi:10.1016/j.tej.2020.106883 · Cited by 29
Performance prediction for convolutional neural networks on edge GPUs
Bouzidi, H., Ouarnoughi, H., Niar, S., & Cadi, A. A. E. (2021) · ACM International Conference on Computing Frontiers (CF) · doi:10.1145/3457388.3458666 · Cited by 35
“Consequently, finding the CNN which gives the best trade-off between accuracy and execution time is a tedious task [1].”
BackgroundDDoS Attack Detection in Vehicular Ad-Hoc Network (VANET) for 5G Networks
Gaurav, A., Gupta, B. B., García‐Peñalvo, F. J., Nedjah, N., & Psannis, K. E. (2021) · Studies in big data · doi:10.1007/978-3-030-85428-7_11 · Cited by 37
Update Compression for Deep Neural Networks on the Edge
Chen, B., Bakhshi, A., Batista, G., Ng, B. W. ‐., & Chin, T.-J. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00347 · Cited by 35
“This is due to the growing popularity of concepts such as AI on the edge [6, 41, 44, 51], which emphasises processing closer to the source of the data to minimise latency and reduce data transfer, and embodied AI [5, 15, 55], which focusses on building intelligent agents that can explore and interact with the environm…”
BackgroundHao, J., Subedi, P., Ramaswamy, L. M., & Kim, I. K. (2023) · ACM Transactions on Internet Technology · doi:10.1145/3546192 · Cited by 35
“Moreover, model compression techniques [27], e.”
BackgroundLee, H., Ryu, D. G., Lee, G., Song, M., Moon, H., Lee, J., Ryu, J., Kang, J., Suh, J. M., Kim, S., et al. (2022) · Advanced Electronic Materials · doi:10.1002/aelm.202200378 · Cited by 30
FEDZIP: A Compression Framework for Communication-Efficient Federated LearningPreprint
Malekijoo, A., Fadaeieslam, M. J., Malekijou, H., Homayounfar, M., Alizadeh-Shabdiz, F., & Rawassizadeh, R. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.01593 · Cited by 90
“As a result, decentralized learning architectures, such as Federated Learning [9–11], on-device machine learning [12,13], and edge computing [14] have been introduced to address this demand.”
BackgroundDistributed artificial intelligence: Taxonomy, review, framework, and reference architecture
Janbi, N. F., Katib, I. A., & Mehmood, R. (2023) · Intelligent Systems with Applications · doi:10.1016/j.iswa.2023.200231 · Cited by 44
A Survey on Hardware Accelerator Design of Deep Learning for Edge Devices
Samanta, A., Hatai, I., & Mal, A. K. (2024) · Wireless Personal Communications · doi:10.1007/s11277-024-11443-2 · Cited by 39
“Protocols for decentralised communication: Until now, centralized training topologies with several endpoints connected to an edge server have only been studied by Chen et al. [61].”
Network for Distributed Intelligence: a Survey and Future Perspectives
Campolo, C., Iera, A., & Molinaro, A. (2023) · IEEE Access · doi:10.1109/access.2023.3280411 · Cited by 44
“Initially proposed for data centers with multiple workers, the same approach has been recently borrowed in the edge computing domain, where several nodes can enforce the aforementioned parallel tasks [28], [29].”
BackgroundFastSecNet: An Efficient Cryptographic Framework for Private Neural Network Inference
Hao, M., Li, H., Chen, H., Xing, P., & Zhang, T. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3262149 · Cited by 27
“various applications like voice assistant [1], medical diagno-”
BackgroundHohman, F., Kery, M. B., Ren, D., & Moritz, D. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642109 · Cited by 50
“Challenges of on-device ML often occur in the context of mobile or edge computing [16, 75, 131].”
BackgroundComputation-Aware Offloading for DNN Inference Tasks in Semantic Communication Assisted MEC Systems
Zheng, G., Wen, M., Ning, Z., & Ding, Z. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3523517 · Cited by 31
“A rapid development of artificial intelligence has led to the widespread use of deep neural network (DNN) applications, e.g., image/video analysis and natural language processing, in a broad range of smart mobile devices [1], [2], [3].”
Artificial Intelligence-Aided Kalman Filters: AI-Augmented Designs for Kalman-Type Algorithms
Shlezinger, N., Revach, G., Ghosh, A., Chatterjee, S., Tang, S., Imbiriba, T., Duník, J., Straka, O., Closas, P., & Eldar, Y. C. (2025) · IEEE Signal Processing Magazine · doi:10.1109/msp.2025.3569395 · Cited by 52
Lightweight and Unobtrusive Data Obfuscation at IoT Edge for Remote Inference
Xu, D., Zheng, M., Jiang, L., Gu, C., Tan, R., & Cheng, P. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2983278 · Cited by 32
Automated Ensemble for Deep Learning Inference on Edge Computing Platforms
Bai, Y., Chen, L., Abdel-Mottaleb, M. S., & Xu, J. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3102945 · Cited by 35
Highly-Optimized Radar-Based Gesture Recognition System with Depthwise Expansion Module
Chmurski, M., Mauro, G., Santra, A., Zubert, M., & Dagasan, G. (2021) · Sensors · doi:10.3390/s21217298 · Cited by 30
“The fundamental principle of edge computing is shifting the computation and communication resources from the cloud to the edge of networks [47], to avoid communication latency, provide a privacy protection capability, and enable a faster response to the end users.”
BackgroundSamikwa, E., Maio, A. D., & Braun, T. I. (2022) · IEEE Consumer Communications and Networking Conference · doi:10.1109/ccnc49033.2022.9700550 · Cited by 36
Survey of Deep Learning Accelerators for Edge and Emerging Computing
Alam, S., Yakopcic, C., Wu, Q., Barnell, M., Khan, S., & Taha, T. M. (2024) · Electronics · doi:10.3390/electronics13152988 · Cited by 38
New Bridge to Cloud: An Ultra-Dense LEO Assisted Green Computation Offloading Approach
Tang, Z., Yu, K. P., Yang, G., Cai, L. X., & Zhou, H. (2022) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2022.3208819 · Cited by 30
“Machine learning, especially deep learning (DL), is an effective technology in wireless resource management and intelligent scheduling, such as intelligent task offload decisions [23].”
BackgroundIntelligent Video Analytics for Human Action Recognition: The State of Knowledge
Kulbacki, M., Segen, J., Chaczko, Z., Rozenblit, J. W., Kulbacki, M., Klempous, R., & Wojciechowski, K. W. (2023) · Sensors · doi:10.3390/s23094258 · Cited by 35
“Action segmentation may include detecting the start and end of action sequences based on changes in common positions or velocities.”
BackgroundScherer, M., Rutishauser, G., Cavigelli, L., & Benini, L. (2021) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2021.3075420 · Cited by 31
Harnessing the Computing Continuum for Urgent Science
Balouek‐Thomert, D., Rodero, I., & Parashar, M. (2020) · ACM SIGMETRICS Performance Evaluation Review · doi:10.1145/3439602.3439618 · Cited by 29
“As data analytics based on AI/ML techniques are becoming an increasingly important component of data-driven application work-� ows, several studies have identi � ed model optimization and ML inference as the main vectors driving the use of resources at the edge [17, 18].”
BackgroundCyber-Physical-Social Systems: An Overview
Reine, R., Juwono, F. H., Sim, Z. A., & Wong, W. K. (2021) · Springer eBooks · doi:10.1007/978-3-030-76387-9_2 · Cited by 33
Na, J., Zhang, H., Lian, J., & Zhang, B. (2022) · Applied Sciences · doi:10.3390/app122010619 · Cited by 30
Machine Learning Based Techniques for Fault Detection in Power Distribution Grid: A Review
Ibitoye, O. T., Onibonoje, M. O., & Dada, J. O. (2022) · doi:10.1109/iconeei55709.2022.9972279 · Cited by 26
“The introduction of extra hidden layers in DL bring about better accuracy [1], [20], [23], [24].”
BackgroundJoint DNN partitioning and task offloading in mobile edge computing via deep reinforcement learning
Zhang, J., Shu-fang, M., Yan, Z., & Huang, J. (2023) · Journal of Cloud Computing Advances Systems and Applications · doi:10.1186/s13677-023-00493-9 · Cited by 33
“…this issue, mobile edge computing (MEC) [5–7] is proposed as an emergent computing mode that places computational and storage resources on edge nodes near MDs [8], enabling compute-intensive DNN-based applications to be executed in a real-time responsive manner, i.e., edge intelligence [1, 9, 10].”
BackgroundMethodEnabling Streaming Analytics in Satellite Edge Computing via Timely Evaluation of Big Data Queries
Xu, Z., Xu, G., Wang, H., Liang, W., Xia, Q., & Wang, S. (2023) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2023.3332333 · Cited by 27
“Low earth orbit (LEO) satellite networks are envisioned as a key technology to construct the satellite Internet for ubiquitous access of ground users on the Earth [4], [38].”
BackgroundGordienko, Y. G., Kochura, Y., Taran, V., Gordienko, N., Rokovyi, O., Alienin, O., & Stirenko, S. (2020) · Advances in computers · doi:10.1016/bs.adcom.2020.10.003 · Cited by 30
Analysis of Edge-Optimized Deep Learning Classifiers for Radar-Based Gesture Recognition
Chmurski, M., Zubert, M., Bierzynski, K., & Santra, A. (2021) · IEEE Access · doi:10.1109/access.2021.3081353 · Cited by 29
“…of edge computing is pushing the computational and communication resources from the cloud to the edge of networks to perform computations [3], [4], [21], [22], thereby avoiding unnecessary communication latencies, providing a privacy protection capability, and enabling faster response for the…”
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.