{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,26]],"date-time":"2026-07-26T03:19:36Z","timestamp":1785035976227,"version":"3.55.0"},"reference-count":150,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2019,8,1]],"date-time":"2019-08-01T00:00:00Z","timestamp":1564617600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2019,8,1]],"date-time":"2019-08-01T00:00:00Z","timestamp":1564617600000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2019,8,1]],"date-time":"2019-08-01T00:00:00Z","timestamp":1564617600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2019,8,1]],"date-time":"2019-08-01T00:00:00Z","timestamp":1564617600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["CNS-1817216"],"award-info":[{"award-number":["CNS-1817216"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. IEEE"],"published-print":{"date-parts":[[2019,8]]},"DOI":"10.1109\/jproc.2019.2921977","type":"journal-article","created":{"date-parts":[[2019,7,15]],"date-time":"2019-07-15T23:43:21Z","timestamp":1563234201000},"page":"1655-1674","source":"Crossref","is-referenced-by-count":1339,"title":["Deep Learning With Edge Computing: A Review"],"prefix":"10.1109","volume":"107","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9923-9027","authenticated-orcid":false,"given":"Jiasi","family":"Chen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xukan","family":"Ran","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/MCI.2018.2840738"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/SEC.2018.00016"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3300061.3300116"},{"key":"ref32","year":"0","journal-title":"Cudnn"},{"key":"ref31","year":"0","journal-title":"Cuda"},{"key":"ref30","year":"0","journal-title":"PyTorch"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/2789168.2790123"},{"key":"ref36","first-page":"91","article-title":"Faster R-Cnn: Towards real-time object detection with region proposal networks","author":"ren","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref35","year":"0","journal-title":"AWS Deeplens"},{"key":"ref34","first-page":"1","article-title":"DeepIoT: Compressing deep neural network structures for sensing systems with a compressor-critic framework","author":"liu","year":"2017","journal-title":"Proc SenSys"},{"key":"ref28","year":"0","journal-title":"Caffe"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"ref29","year":"0","journal-title":"Caffe"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2018.8485905"},{"key":"ref22","article-title":"MnasNet: Platform-aware neural architecture search for mobile","author":"tan","year":"2018","journal-title":"arXiv 1807 11626"},{"key":"ref21","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"howard","year":"2017","journal-title":"arXiv 1704 04861"},{"key":"ref24","year":"0","journal-title":"TensorFlow&#x2122;"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3211332.3211336"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2807412"},{"key":"ref26","year":"0","journal-title":"Tensorflowlite"},{"key":"ref100","first-page":"1","article-title":"Very deep convolutional networks for large-scale image recognition","volume":"abs 1409 1556","author":"simonyan","year":"2014","journal-title":"CoRR"},{"key":"ref25","first-page":"1","article-title":"TensorFlow: Large-scale machine learning on heterogeneous distributed systems","volume":"abs 1603 4467","author":"abadi","year":"2016","journal-title":"CoRR"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2018.23204"},{"key":"ref51","first-page":"1","article-title":"Cellular network traffic scheduling with deep reinforcement learning","author":"chinchali","year":"2018","journal-title":"Proc 32nd AAAI Conf Artif Intell"},{"key":"ref150","doi-asserted-by":"crossref","first-page":"265","DOI":"10.1007\/11681878_14","article-title":"Calibrating noise to sensitivity in private data analysis","author":"dwork","year":"2006","journal-title":"Proc Theory Cryptogr Conf"},{"key":"ref146","doi-asserted-by":"publisher","DOI":"10.1145\/2999572.2999602"},{"key":"ref147","doi-asserted-by":"publisher","DOI":"10.1109\/INFCOM.2012.6195689"},{"key":"ref148","doi-asserted-by":"publisher","DOI":"10.1145\/2999572.2999604"},{"key":"ref149","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2017.41"},{"key":"ref59","first-page":"1935","article-title":"Resource-efficient machine learning in 2 KB RAM for the Internet of Things","volume":"70","author":"kumar","year":"2017","journal-title":"Proc 34th Int Conf Mach Learn"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/MCOM.2017.1700168"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2013.257"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1145\/3038912.3052577"},{"key":"ref55","first-page":"1","article-title":"A deep reinforcement learning-based framework for content caching","volume":"abs 1712 8132","author":"zhong","year":"2017","journal-title":"CoRR"},{"key":"ref54","first-page":"1","article-title":"Distributed deep learning at the edge: A novel proactive and cooperative caching framework for mobile edge networks","volume":"abs 1812 5374","author":"saputra","year":"2018","journal-title":"CoRR"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/MNET.2018.1800109"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/TETC.2014.2369958"},{"key":"ref40","year":"0","journal-title":"Deep Learning for Siri&#x2019;s Voice On-Device Deep Mixture Density Networks for Hybrid Unit Selection Synthesis"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/MPRV.2009.82"},{"key":"ref3","author":"jeans","year":"2019","journal-title":"Related&#x2019;s Hudson Yards Smart City or Surveillance City?"},{"key":"ref6","author":"blog","year":"0","journal-title":"Edge Computing at Chick-Fil-A"},{"key":"ref5","year":"0","journal-title":"AT&T Multi-Access Edge Computing"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2016.2579198"},{"key":"ref49","first-page":"943","article-title":"Intrusion detection with neural networks","author":"ryan","year":"1998","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref7","volume":"1","author":"goodfellow","year":"2016","journal-title":"Deep Learning"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2017.2745201"},{"key":"ref46","year":"2017","journal-title":"Hey Siri An On-Device DNN-Powered Voice Trigger for Apple&#x2019;s Personal Assistant"},{"key":"ref45","year":"0","journal-title":"Pagespeed Insights Improve Server Response Time"},{"key":"ref48","year":"0","journal-title":"Google&#x2019;s Pixel Buds are no Match for Professional Interpreters"},{"key":"ref47","first-page":"9017","article-title":"FastGRNN: A fast, accurate, stable and tiny kilobyte sized gated recurrent neural network","author":"kusupati","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref42","article-title":"Google&#x2019;s neural machine translation system: Bridging the gap between human and machine translation","author":"wu","year":"2016","journal-title":"arXiv 1609 08144"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N16-1030"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.279"},{"key":"ref43","year":"0","journal-title":"Google duplex An AI system for accomplishing real-world tasks over the phone"},{"key":"ref127","first-page":"629","article-title":"Gaia: Geo-distributed machine learning approaching LAN speeds","author":"hsieh","year":"2017","journal-title":"Proc USENIX Symp Netw Syst Design Implem (NSDI)"},{"key":"ref126","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2018.8486403"},{"key":"ref125","article-title":"Communication-efficient learning of deep networks from decentralized data","author":"mcmahan","year":"2016","journal-title":"arXiv 1602 05629"},{"key":"ref124","first-page":"685","article-title":"Deep learning with elastic averaging SGD","author":"zhang","year":"2015","journal-title":"Advances in neural information processing systems"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021745"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1145\/3240765.3243473"},{"key":"ref129","first-page":"1","article-title":"Deep gradient compression: Reducing the communication bandwidth for distributed training","volume":"abs 1712 1887","author":"lin","year":"2017","journal-title":"CoRR"},{"key":"ref71","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","author":"han","year":"2015","journal-title":"arXiv 1510 00149 [cs]"},{"key":"ref128","first-page":"1","article-title":"Large scale distributed neural network training through online distillation","volume":"abs 1804 3235","author":"anil","year":"2018","journal-title":"CoRR"},{"key":"ref70","article-title":"SqueezeNet: Alexnet-level accuracy with 50x fewer parameters and <0.5 MB model size","author":"iandola","year":"2016","journal-title":"arXiv 1602 07360"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR.2016.7900006"},{"key":"ref130","doi-asserted-by":"publisher","DOI":"10.1109\/NCA.2017.8171350"},{"key":"ref77","first-page":"389","article-title":"On-demand deep model compression for mobile devices: A usage-driven model selection framework","author":"yao","year":"2018","journal-title":"Proc MobiSys"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1145\/2994551.2994564"},{"key":"ref75","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"2015","journal-title":"ArXiv 1503 02531"},{"key":"ref133","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2018.00023"},{"key":"ref134","doi-asserted-by":"publisher","DOI":"10.1145\/2976749.2978318"},{"key":"ref131","article-title":"Gossip training for deep learning","author":"blot","year":"2016","journal-title":"arXiv 1611 09726"},{"key":"ref78","first-page":"82","article-title":"DeepMon: Mobile GPU-based deep learning framework for continuous vision applications","author":"loc huynh","year":"2017","journal-title":"Proc ACM MOBISYS"},{"key":"ref132","article-title":"How to scale distributed deep learning?","author":"jin","year":"2016","journal-title":"arXiv 1611 04581"},{"key":"ref79","year":"0","journal-title":"Edge TPU"},{"key":"ref136","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2017.12"},{"key":"ref135","article-title":"Privacy-preserving machine learning through data obfuscation","author":"zhang","year":"2018","journal-title":"arXiv 1807 01860"},{"key":"ref138","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(98)00116-6"},{"key":"ref137","doi-asserted-by":"publisher","DOI":"10.1109\/SEC.2018.00014"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2858384"},{"key":"ref139","first-page":"1","article-title":"Accurate, large minibatch SGD: Training imagenet in 1 hour","volume":"abs 1706 2677","author":"goyal","year":"2017","journal-title":"CoRR"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1145\/3229625.3229629"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2018.2844341"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00559"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1145\/3097895.3097896"},{"key":"ref140","doi-asserted-by":"publisher","DOI":"10.1109\/NCA.2017.8171350"},{"key":"ref65","author":"jindal","year":"2018","journal-title":"Enabling full body AR with Mask R-CNN2Go"},{"key":"ref141","doi-asserted-by":"publisher","DOI":"10.1145\/2462456.2464448"},{"key":"ref66","author":"lavalle","year":"2016","journal-title":"Virtual Reality"},{"key":"ref142","doi-asserted-by":"publisher","DOI":"10.1145\/3081333.3081347"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1145\/3132211.3134458"},{"key":"ref143","first-page":"12","article-title":"You can teach elephants to dance: Agile VM handoff for edge computing","author":"ha","year":"2017","journal-title":"Proc Symp Comput Commun"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1145\/2594368.2594383"},{"key":"ref144","doi-asserted-by":"publisher","DOI":"10.1145\/3132211.3134460"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/MC.2017.9"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"ref145","doi-asserted-by":"publisher","DOI":"10.1145\/3132211.3134445"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref109","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2017.226"},{"key":"ref95","first-page":"29","article-title":"Mainstream: Dynamic stem-sharing for multi-tenant video processing","author":"jiang","year":"2018","journal-title":"Proc Usenix Ann Technical Conf (Usenix '99)"},{"key":"ref108","year":"0","journal-title":"Arcore Overview"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1145\/3230543.3230574"},{"key":"ref107","doi-asserted-by":"publisher","DOI":"10.23919\/DATE.2017.7927211"},{"key":"ref93","first-page":"377","article-title":"Live video analytics at scale with approximation and delay-tolerance","author":"zhang","year":"2017","journal-title":"Proc USENIX NSDI"},{"key":"ref106","doi-asserted-by":"publisher","DOI":"10.1109\/MNET.2018.1700202"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1109\/TSC.2017.2662008"},{"key":"ref105","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.220"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1145\/2809695.2809711"},{"key":"ref104","doi-asserted-by":"publisher","DOI":"10.1145\/2906388.2906396"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1109\/MCOM.2018.1701277"},{"key":"ref103","first-page":"1","article-title":"Deep residual learning for image recognition","volume":"abs 1512 3385","author":"he","year":"2015","journal-title":"CoRR"},{"key":"ref102","first-page":"1","article-title":"Going deeper with convolutions","volume":"abs 1409 4842","author":"szegedy","year":"2014","journal-title":"CoRR"},{"key":"ref111","doi-asserted-by":"publisher","DOI":"10.1145\/3097895.3097903"},{"key":"ref112","doi-asserted-by":"publisher","DOI":"10.1145\/1814433.1814441"},{"key":"ref110","first-page":"1","article-title":"BranchyNet: Fast inference via early exiting from deep neural networks","volume":"abs 1709 1686","author":"teerapittayanon","year":"2017","journal-title":"CoRR"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref99","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref97","first-page":"1","article-title":"Bidirectional LSTM-CRF models for sequence tagging","volume":"abs 1508 1991","author":"huang","year":"2015","journal-title":"CoRR"},{"key":"ref10","first-page":"1","article-title":"Precog: Prefetching for image recognition applications at the edge","author":"drolia","year":"2017","journal-title":"Proc ACM\/IEEE Symp Edge Comput"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/MNET.2018.1700202"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3081333.3081360"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3093336.3037698"},{"key":"ref14","doi-asserted-by":"crossref","first-page":"1310","DOI":"10.1145\/2810103.2813687","article-title":"Privacy-preserving deep learning","author":"shokri","year":"2015","journal-title":"Proc 22nd ACM SIGSAC Conf Comput Commun Secur"},{"key":"ref15","first-page":"1","article-title":"An overview of gradient descent optimization algorithms","volume":"abs 1609 4747","author":"ruder","year":"2016","journal-title":"CoRR"},{"key":"ref118","doi-asserted-by":"publisher","DOI":"10.1145\/3195970.3196023"},{"key":"ref16","first-page":"311","article-title":"BLEU: A method for automatic evaluation of machine translation","author":"papineni","year":"2002","journal-title":"Proc Annual Meeting of the Assoc Computational Linguistics"},{"key":"ref82","author":"ovtcharov","year":"0","journal-title":"Accelerating deep convolutional neural networks using specialized hardware"},{"key":"ref117","doi-asserted-by":"publisher","DOI":"10.1145\/3133956.3134056"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.351"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1145\/2996864"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.690"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2018.00098"},{"key":"ref119","first-page":"1651","article-title":"GAZELLE: A low latency framework for secure neural network inference","author":"juvekar","year":"2018","journal-title":"Proc 27th USENIX Security Symp (USENIX Security)"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3123266.3123389"},{"key":"ref83","year":"0","journal-title":"VPU"},{"key":"ref114","doi-asserted-by":"publisher","DOI":"10.1145\/1999995.2000000"},{"key":"ref113","doi-asserted-by":"publisher","DOI":"10.1145\/3132211.3134459"},{"key":"ref116","first-page":"201","article-title":"CryptoNets: Applying neural networks to encrypted data with high throughput and accuracy","author":"gilad-bachrach","year":"2016","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1145\/2872887.2750389"},{"key":"ref115","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3220106"},{"key":"ref120","doi-asserted-by":"publisher","DOI":"10.1145\/3196494.3196522"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2017.2761740"},{"key":"ref121","first-page":"1223","article-title":"Large scale distributed deep networks","author":"dean","year":"2012","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref122","doi-asserted-by":"publisher","DOI":"10.1109\/SEC.2016.38"},{"key":"ref123","first-page":"1","article-title":"Revisiting distributed synchronous SGD","volume":"abs 1604 981","author":"chen","year":"2016","journal-title":"CoRR"},{"key":"ref85","year":"0","journal-title":"NVIDIA EGX Edge Computing Platform"},{"key":"ref86","year":"0","journal-title":"Qualcomm Neural Processing SDK for AI"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1145\/3089801.3089805"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1109\/IPSN.2016.7460664"}],"container-title":["Proceedings of the IEEE"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielaam\/5\/8789751\/8763885-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/5\/8789751\/08763885.pdf?arnumber=8763885","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,7,13]],"date-time":"2022-07-13T20:58:45Z","timestamp":1657745925000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/8763885\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,8]]},"references-count":150,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/jproc.2019.2921977","relation":{},"ISSN":["0018-9219","1558-2256"],"issn-type":[{"value":"0018-9219","type":"print"},{"value":"1558-2256","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,8]]}}}