[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_279":280},"routes/_layout",{"_11":206},"routes/entity.$type.$id",{"_11":12,"_13":14,"_15":16,"_17":18,"_19":20,"_21":22,"_23":24,"_25":26,"_27":28},"dehydratedState",{"_73":74,"_75":76},"type","paper","id","paper_W2980856918","section","technologies","entity",{"_15":16,"_29":30,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":38,"_39":-5,"_40":41,"_42":43,"_44":45,"_46":47,"_48":-5,"_49":50,"_51":30,"_52":53,"_54":55,"_56":57,"_58":59,"_60":61,"_62":-5,"_63":64,"_65":-5,"_66":67,"_68":30,"_69":-5},"canonical","https://www.hugmap.com/entity/paper/paper_W2980856918/technologies","siteBaseUrl","https://www.hugmap.com","sectionCount",1,"jsonLd","{\"@context\":\"https://schema.org\",\"@graph\":[{\"name\":\"Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing\",\"url\":\"https://www.hugmap.com/entity/paper/paper_W2980856918/technologies\",\"@type\":\"ScholarlyArticle\",\"headline\":\"Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing\",\"datePublished\":\"2019-10-18\",\"isPartOf\":{\"@type\":\"Periodical\",\"name\":\"IEEE Transactions on Wireless Communications\"}},{\"@type\":\"CollectionPage\",\"name\":\"Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing 相关技术\",\"description\":\"Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing 在 HugMap 中的相关技术，共 1 条，含结构化关联信息与详情链接。\",\"url\":\"https://www.hugmap.com/entity/paper/paper_W2980856918/technologies\",\"inLanguage\":\"zh-CN\"},{\"@type\":\"BreadcrumbList\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"首页\",\"item\":\"https://www.hugmap.com/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"论文\",\"item\":\"https://www.hugmap.com/search?type=paper\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing\",\"item\":\"https://www.hugmap.com/entity/paper/paper_W2980856918\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"相关技术\",\"item\":\"https://www.hugmap.com/entity/paper/paper_W2980856918/technologies\"}]}]}","name","Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing","nameZh","aliases","description","descriptionZh","status","active","externalIds","{\"openalex\":\"W2980856918\",\"doi\":\"10.1109/twc.2019.2946140\"}","properties","source","openalex","createdAt","2026-05-14T20:02:51.429246588","updatedAt","2026-05-20T20:28:38.211656395","doi","10.1109/twc.2019.2946140","arxivId","openalexId","W2980856918","title","abstractText","As a key technology of enabling Artificial Intelligence (AI) applications in 5G era, Deep Neural Networks (DNNs) have quickly attracted widespread attention. However, it is challenging to run computation-intensive DNN-based tasks on mobile devices due to the limited computation resources. What’s worse, traditional cloud-assisted DNN inference is heavily hindered by the significant wide-area network latency, leading to poor real-time performance as well as low quality of user experience. To address these challenges, in this paper, we propose Edgent , a framework that leverages edge computing for DNN collaborative inference through device-edge synergy. Edgent exploits two design knobs: (1) DNN partitioning that adaptively partitions computation between device and edge for purpose of coordinating the powerful cloud resource and the proximal edge resource for real-time DNN inference; (2) DNN right-sizing that further reduces computing latency via early exiting inference at an appropriate intermediate DNN layer. In addition, considering the potential network fluctuation in real-world deployment, Edgent is properly design to specialize for both static and dynamic network environment. Specifically, in a static environment where the bandwidth changes slowly, Edgent derives the best configurations with the assist of regression-based prediction models, while in a dynamic environment where the bandwidth varies dramatically, Edgent generates the best execution plan through the online change point detection algorithm that maps the current bandwidth state to the optimal configuration. We implement Edgent prototype based on the Raspberry Pi and the desktop PC and the extensive experimental evaluations demonstrate Edgent ’s effectiveness in enabling on-demand low-latency edge intelligence.","publishedDate","2019-10-18","publishedYear",2019,"venue","IEEE Transactions on Wireless Communications","citationCount",844,"authors","keywords",[70,71,72],"pdfUrl","topVenue",false,"displayName","displayDescription","IoT and Edge/Fog Computing","Advanced Neural Network Applications","Blockchain Technology Applications and Security","mutations",[],"queries",[77],{"_78":79,"_80":81,"_82":83,"_84":85},"dehydratedAt",1780619762052,"state",{"_87":88,"_89":26,"_90":79,"_91":-5,"_92":93,"_94":93,"_95":93,"_96":-5,"_97":-5,"_98":67,"_35":99,"_100":101},"queryKey",[86,19,14,16],"queryHash","[\"portal\",\"entity\",\"paper\",\"paper_W2980856918\"]","portal","data",{"_19":20,"_102":103,"_104":105,"_106":107,"_108":109,"_18":110,"_62":111},"dataUpdateCount","dataUpdatedAt","error","errorUpdateCount",0,"errorUpdatedAt","fetchFailureCount","fetchFailureReason","fetchMeta","isInvalidated","success","fetchStatus","idle","entityType","Paper","taxonomies",[165,166,167,168],"events",[],"articles",[],[150],[112,113,114,115],{"_15":144,"_29":145,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":146,"_39":-5,"_40":41,"_42":147,"_44":148,"_123":149,"_51":-5,"_124":-5,"_125":-5,"_126":-5,"_60":-5,"_127":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_68":145,"_69":-5},{"_15":138,"_29":139,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":140,"_39":-5,"_40":41,"_42":141,"_44":142,"_123":143,"_51":-5,"_124":-5,"_125":-5,"_126":-5,"_60":-5,"_127":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_68":139,"_69":-5},{"_15":132,"_29":133,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":134,"_39":-5,"_40":41,"_42":135,"_44":136,"_123":137,"_51":-5,"_124":-5,"_125":-5,"_126":-5,"_60":-5,"_127":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_68":133,"_69":-5},{"_15":116,"_29":117,"_31":-5,"_32":118,"_33":-5,"_34":-5,"_35":36,"_37":119,"_39":-5,"_40":120,"_42":121,"_44":122,"_123":-5,"_51":-5,"_124":-5,"_125":-5,"_126":-5,"_60":-5,"_127":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_68":117,"_69":-5},"person_A5100738931","En Li",[],"{\"openalex\":\"A5100738931\"}","openalex,graphrag","2026-05-14T04:22:15.379243872","2026-05-29T09:35:23.892047837","orcid","affiliation","researchArea","paperCount","homepage","linkedin","twitter","github","hindex","person_A5055161955","Liekang Zeng","{\"openalex\":\"A5055161955\",\"orcid\":\"0000-0003-4800-8768\"}","2026-05-14T04:22:16.352954473","2026-05-22T02:35:08.219852732","0000-0003-4800-8768","person_A5100760218","Zhi Zhou","{\"openalex\":\"A5100760218\",\"orcid\":\"0000-0002-0987-9344\"}","2026-05-14T04:22:15.293159313","2026-05-22T02:35:08.231227136","0000-0002-0987-9344","person_A5100385692","Xu Chen","{\"openalex\":\"A5100385692\",\"orcid\":\"0000-0001-9943-6020\"}","2026-05-14T03:13:50.271412984","2026-05-21T17:08:56.269255728","0000-0001-9943-6020",{"_15":151,"_29":71,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":152,"_39":-5,"_40":153,"_42":154,"_44":155,"_156":157,"_158":159,"_160":-5,"_161":162,"_163":-5,"_126":-5,"_164":-5,"_68":71,"_69":-5},"tech_oa_T10036","{\"taxonomy_id\":\"tax_rd_T10036\"}","openalex_topics","2026-05-17T09:00:00.301982826","2026-05-17T09:00:00.301985682","techType","architecture","domain","Physical Sciences","proposedYear","maturityLevel","growing","wikipediaUrl","patentCount",{"_15":199,"_29":200,"_31":-5,"_32":-5,"_33":201,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":205,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":200,"_69":201},{"_15":193,"_29":71,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":194,"_39":-5,"_40":41,"_42":195,"_44":196,"_173":174,"_175":-5,"_176":177,"_178":197,"_180":198,"_182":93,"_183":184,"_185":93,"_186":-5,"_68":71,"_69":-5},{"_15":187,"_29":70,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":188,"_39":-5,"_40":41,"_42":189,"_44":190,"_173":174,"_175":-5,"_176":177,"_178":191,"_180":192,"_182":93,"_183":184,"_185":93,"_186":-5,"_68":70,"_69":-5},{"_15":169,"_29":72,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":170,"_39":-5,"_40":41,"_42":171,"_44":172,"_173":174,"_175":-5,"_176":177,"_178":179,"_180":181,"_182":93,"_183":184,"_185":93,"_186":-5,"_68":72,"_69":-5},"tax_rd_T10270","{\"openalex_topic\":\"T10270\"}","2026-05-14T01:25:05.372191624","2026-05-23T12:36:17.219717842","taxonomyType","RESEARCH_DOMAIN","code","level",3,"parentId","tax_rd_1710","path","Physical Sciences/Computer Science/Information Systems/Blockchain Technology Applications and Security","sortOrder","isLeaf",true,"entityCount","children","tax_rd_T10273","{\"openalex_topic\":\"T10273\"}","2026-05-14T00:51:58.346333431","2026-05-23T12:36:17.211721119","tax_rd_1705","Physical Sciences/Computer Science/Computer Networks and Communications/IoT and Edge/Fog Computing","tax_rd_T10036","{\"openalex_topic\":\"T10036\"}","2026-05-14T00:27:13.178912305","2026-05-23T12:38:48.227193842","tax_rd_1707","Physical Sciences/Computer Science/Computer Vision and Pattern Recognition/Advanced Neural Network Applications","tax_ind_security","网络安全","Cybersecurity","manual","INDUSTRY","tax_ind_root","科技行业/网络安全",{"_73":207,"_75":208},[],[209],{"_78":210,"_80":211,"_82":212,"_84":213},1780619761333,{"_87":215,"_89":26,"_90":210,"_91":-5,"_92":93,"_94":93,"_95":93,"_96":-5,"_97":-5,"_98":67,"_35":99,"_100":101},[86,214],"[\"portal\",\"industries\"]","industries",[216,217,218,219,220,221,222,223,224,225,226,227],{"_15":272,"_29":273,"_31":-5,"_32":274,"_33":275,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":276,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":273,"_69":275},{"_15":268,"_29":269,"_31":-5,"_32":-5,"_33":270,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":271,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":269,"_69":270},{"_15":262,"_29":263,"_31":-5,"_32":264,"_33":265,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":266,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":263,"_69":265},{"_15":199,"_29":200,"_31":-5,"_32":-5,"_33":201,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":205,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":200,"_69":201},{"_15":258,"_29":259,"_31":-5,"_32":-5,"_33":260,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":261,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":259,"_69":260},{"_15":252,"_29":253,"_31":-5,"_32":254,"_33":255,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":256,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":253,"_69":255},{"_15":248,"_29":249,"_31":-5,"_32":-5,"_33":250,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":251,"_182":93,"_183":184,"_185":93,"_186":-5,"_68":249,"_69":250},{"_15":244,"_29":245,"_31":-5,"_32":-5,"_33":246,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":247,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":245,"_69":246},{"_15":240,"_29":241,"_31":-5,"_32":-5,"_33":242,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":243,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":241,"_69":242},{"_15":236,"_29":237,"_31":-5,"_32":-5,"_33":238,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":239,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":237,"_69":238},{"_15":232,"_29":233,"_31":-5,"_32":-5,"_33":234,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":235,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":233,"_69":234},{"_15":228,"_29":229,"_31":-5,"_32":-5,"_33":230,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":202,"_42":-5,"_44":-5,"_173":203,"_175":-5,"_176":26,"_178":204,"_180":231,"_182":93,"_183":67,"_185":93,"_186":-5,"_68":229,"_69":230},"tax_ind_smartphone","智能手机","Smartphone","科技行业/智能手机","tax_ind_robot","机器人","Robotics","科技行业/机器人","tax_ind_wearable","智能穿戴","Smart Wearables","科技行业/智能穿戴","tax_ind_energy","新能源与智能制造","New Energy & Smart Manufacturing","科技行业/新能源与智能制造","tax_ind_biotech","生物科技","Biotech","科技行业/生物科技","tax_ind_quantum","量子计算","Quantum Computing","科技行业/量子计算","tax_ind_blockchain","区块链",[257],"Blockchain","科技行业/区块链","Web3","tax_ind_enterprise","企业软件","Enterprise Software","科技行业/企业软件","tax_ind_semi","半导体",[267],"Semiconductor","科技行业/半导体","芯片","tax_ind_cloud","云计算","Cloud Computing","科技行业/云计算","tax_ind_ai","人工智能",[277,278],"Artificial Intelligence","科技行业/人工智能","AI","AI 技术","env",{"_281":24},"SITE_BASE_URL"]

Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing 论文

Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing · 相关技术

相关技术