{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T19:24:17Z","timestamp":1772997857830,"version":"3.50.1"},"reference-count":40,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,8,3]],"date-time":"2023-08-03T00:00:00Z","timestamp":1691020800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,8,3]],"date-time":"2023-08-03T00:00:00Z","timestamp":1691020800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Cloud Comp"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>As Artificial Intelligence (AI) becomes increasingly prevalent, Deep Neural Networks (DNNs) have become a crucial tool for developing and advancing AI applications. Considering limited computing and energy resources on mobile devices (MDs), it is a challenge to perform compute-intensive DNN tasks on MDs. To attack this challenge, mobile edge computing (MEC) provides a viable solution through DNN partitioning and task offloading. However, as the communication conditions between different devices change over time, DNN partitioning on different devices must also change synchronously. This is a dynamic process, which aggravates the complexity of DNN partitioning. In this paper, we delve into the issue of jointly optimizing energy and delay for DNN partitioning and task offloading in a dynamic MEC scenario where each MD and the server adopt the pre-trained DNNs for task inference. Taking advantage of the characteristics of DNN, we first propose a strategy for layered partitioning of DNN tasks to divide the task of each MD into subtasks that can be either processed on the MD or offloaded to the server for computation. Then, we formulate the trade-off between energy and delay as a joint optimization problem, which is further represented as a Markov decision process (MDP). To solve this, we design a DNN partitioning and task offloading (DPTO) algorithm utilizing deep reinforcement learning (DRL), which enables MDs to make optimal offloading decisions. Finally, experimental results demonstrate that our algorithm outperforms existing non-DRL and DRL algorithms with respect to processing delay and energy consumption, and can be applied to different DNN types.<\/jats:p>","DOI":"10.1186\/s13677-023-00493-9","type":"journal-article","created":{"date-parts":[[2023,8,3]],"date-time":"2023-08-03T04:02:32Z","timestamp":1691035352000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":21,"title":["Joint DNN partitioning and task offloading in mobile edge computing via deep reinforcement learning"],"prefix":"10.1186","volume":"12","author":[{"given":"Jianbing","family":"Zhang","sequence":"first","affiliation":[]},{"given":"Shufang","family":"Ma","sequence":"additional","affiliation":[]},{"given":"Zexiao","family":"Yan","sequence":"additional","affiliation":[]},{"given":"Jiwei","family":"Huang","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,8,3]]},"reference":[{"issue":"8","key":"493_CR1","doi-asserted-by":"publisher","first-page":"1655","DOI":"10.1109\/JPROC.2019.2921977","volume":"107","author":"J Chen","year":"2019","unstructured":"Chen J, Ran X (2019) Deep learning with edge computing: A review. Proc IEEE 107(8):1655\u20131674","journal-title":"Proc IEEE"},{"key":"493_CR2","doi-asserted-by":"publisher","unstructured":"Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A (2015) Going deeper with convolutions. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). p 1\u20139. https:\/\/doi.org\/10.1109\/CVPR.2015.7298594","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"493_CR3","doi-asserted-by":"publisher","unstructured":"Wang D, Nyberg E (2015) A long short-term memory model for answer sentence selection in question answering. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers). Association for Computational Linguistics, Beijing, China. p 707\u2013712. https:\/\/doi.org\/10.3115\/v1\/P15-2116","DOI":"10.3115\/v1\/P15-2116"},{"key":"493_CR4","doi-asserted-by":"crossref","unstructured":"Chen Z, Hu J, Chen X, Hu J, Zheng X, Min G (2020) Computation offloading and task scheduling for dnn-based applications in cloud-edge computing. IEEE Access 8:115537\u2013115547","DOI":"10.1109\/ACCESS.2020.3004509"},{"issue":"4","key":"493_CR5","doi-asserted-by":"publisher","first-page":"2322","DOI":"10.1109\/COMST.2017.2745201","volume":"19","author":"Y Mao","year":"2017","unstructured":"Mao Y, You C, Zhang J, Huang K, Letaief KB (2017) A survey on mobile edge computing: The communication perspective. IEEE Commun Surv Tutor 19(4):2322\u20132358","journal-title":"IEEE Commun Surv Tutor"},{"issue":"3","key":"493_CR6","doi-asserted-by":"publisher","first-page":"1628","DOI":"10.1109\/COMST.2017.2682318","volume":"19","author":"P Mach","year":"2017","unstructured":"Mach P, Becvar Z (2017) Mobile edge computing: A survey on architecture and computation offloading. IEEE Commun Surv Tutor 19(3):1628\u20131656","journal-title":"IEEE Commun Surv Tutor"},{"issue":"3","key":"493_CR7","doi-asserted-by":"publisher","first-page":"2038","DOI":"10.1109\/JIOT.2019.2960631","volume":"7","author":"Z Xiao","year":"2020","unstructured":"Xiao Z, Dai X, Jiang H, Wang D, Chen H, Yang L, Zeng F (2020) Vehicular task offloading via heat-aware MEC cooperation using game-theoretic method. IEEE Internet Things J 7(3):2038\u20132052","journal-title":"IEEE Internet Things J"},{"issue":"8","key":"493_CR8","doi-asserted-by":"publisher","first-page":"1584","DOI":"10.1109\/JPROC.2019.2922285","volume":"107","author":"L Lin","year":"2019","unstructured":"Lin L, Liao X, Jin H, Li P (2019) Computation offloading toward edge computing. Proc IEEE 107(8):1584\u20131607","journal-title":"Proc IEEE"},{"issue":"1","key":"493_CR9","first-page":"447","volume":"19","author":"E Li","year":"2020","unstructured":"Li E, Zeng L, Zhou Z, Chen X (2020) Edge AI: On-demand accelerating deep neural network inference via edge computing. IEEE Trans Vis Comput Graph 19(1):447\u2013457","journal-title":"IEEE Trans Vis Comput Graph"},{"issue":"11","key":"493_CR10","doi-asserted-by":"publisher","first-page":"1778","DOI":"10.1109\/JPROC.2021.3119950","volume":"109","author":"D Xu","year":"2021","unstructured":"Xu D, Li T, Li Y, Su X, Tarkoma S, Jiang T, Crowcroft J, Hui P (2021) Edge intelligence: Empowering intelligence to the edge of network. Proc IEEE 109(11):1778\u20131837","journal-title":"Proc IEEE"},{"issue":"12","key":"493_CR11","doi-asserted-by":"publisher","first-page":"9511","DOI":"10.1109\/JIOT.2020.3010258","volume":"8","author":"X Tang","year":"2021","unstructured":"Tang X, Chen X, Zeng L, Yu S, Chen L (2021) Joint multiuser DNN partitioning and computational resource allocation for collaborative edge intelligence. IEEE Internet Things J 8(12):9511\u20139522","journal-title":"IEEE Internet Things J"},{"key":"493_CR12","doi-asserted-by":"publisher","unstructured":"Dong F, Wang H, Shen D, Huang Z, He Q, Zhang J, Wen L, Zhang T (2022) Multi-exit DNN inference acceleration based on multi-dimensional optimization for edge intelligence. IEEE Trans Mob Comput 1. https:\/\/doi.org\/10.1109\/TMC.2022.3172402","DOI":"10.1109\/TMC.2022.3172402"},{"issue":"4","key":"493_CR13","doi-asserted-by":"publisher","first-page":"3973","DOI":"10.1109\/TNSM.2021.3116665","volume":"18","author":"C Dong","year":"2021","unstructured":"Dong C, Hu S, Chen X, Wen W (2021) Joint optimization with DNN partitioning and resource allocation in mobile edge computing. IEEE Trans Netw Serv Manag 18(4):3973\u20133986","journal-title":"IEEE Trans Netw Serv Manag"},{"issue":"4","key":"493_CR14","doi-asserted-by":"publisher","first-page":"2435","DOI":"10.1109\/TMC.2021.3114193","volume":"22","author":"M Gao","year":"2023","unstructured":"Gao M, Shen R, Shi L, Qi W, Li J, Li Y (2023) Task partitioning and offloading in DNN-task enabled mobile edge computing networks. IEEE Trans Mob Comput 22(4):2435\u20132445","journal-title":"IEEE Trans Mob Comput"},{"key":"493_CR15","first-page":"71","volume":"20","author":"W Li","year":"2020","unstructured":"Li W (2020) Geoai: Where machine learning and big data converge in giscience. J Spat Inf Sci 20:71\u201377","journal-title":"J Spat Inf Sci"},{"key":"493_CR16","doi-asserted-by":"crossref","unstructured":"Li W, Batty M, Goodchild MF (2020) Real-time GIS for smart cities. J Geog Inf Sci 34(2):311\u2013324","DOI":"10.1080\/13658816.2019.1673397"},{"issue":"8","key":"493_CR17","doi-asserted-by":"publisher","first-page":"1738","DOI":"10.1109\/JPROC.2019.2918951","volume":"107","author":"Z Zhou","year":"2019","unstructured":"Zhou Z, Chen X, Li E, Zeng L, Luo K, Zhang J (2019) Edge intelligence: Paving the last mile of artificial intelligence with edge computing. Proc IEEE 107(8):1738\u20131762","journal-title":"Proc IEEE"},{"key":"493_CR18","doi-asserted-by":"crossref","unstructured":"Kang Y, Hauswald J, Gao C, Rovinski A, Mudge T, Mars J, Tang L (2017) Neurosurgeon: Collaborative intelligence between the cloud and mobile edge. SIGARCH Comput Archit News 45(1):615\u2013629","DOI":"10.1145\/3093337.3037698"},{"issue":"2","key":"493_CR19","doi-asserted-by":"publisher","first-page":"1491","DOI":"10.1109\/JIOT.2021.3088607","volume":"9","author":"G Zhang","year":"2022","unstructured":"Zhang G, Ni S, Zhao P (2022) Learning-based joint optimization of energy delay and privacy in multiple-user edge-cloud collaboration MEC systems. IEEE Internet Things J 9(2):1491\u20131502","journal-title":"IEEE Internet Things J"},{"issue":"10","key":"493_CR20","doi-asserted-by":"publisher","first-page":"9241","DOI":"10.1109\/JIOT.2020.2981338","volume":"7","author":"W He","year":"2020","unstructured":"He W, Guo S, Guo S, Qiu X, Qi F (2020) Joint DNN partition deployment and resource allocation for delay-sensitive deep learning inference in IoT. IEEE Internet Things J 7(10):9241\u20139254","journal-title":"IEEE Internet Things J"},{"key":"493_CR21","doi-asserted-by":"publisher","DOI":"10.1016\/j.buildenv.2022.109513","author":"K Li","year":"2022","unstructured":"Li K, Zhao J, Hu J et al (2022) Dynamic energy efficient task offloading and resource allocation for noma-enabled IoT in smart buildings and environment. Build Environ. https:\/\/doi.org\/10.1016\/j.buildenv.2022.109513","journal-title":"Build Environ"},{"key":"493_CR22","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2022.3223119","author":"Y Chen","year":"2022","unstructured":"Chen Y, Zhao J, Wu Y et al (2022) Qoe-aware decentralized task offloading and resource allocation for end-edge-cloud systems: A game-theoretical approach. IEEE Trans Mob Comput. https:\/\/doi.org\/10.1109\/TMC.2022.3223119","journal-title":"IEEE Trans Mob Comput"},{"key":"493_CR23","doi-asserted-by":"publisher","DOI":"10.1109\/JSYST.2023.3249217","author":"J Huang","year":"2023","unstructured":"Huang J, Wan J, Lv B et al (2023) Joint computation offloading and resource allocation for edge-cloud collaboration in internet of vehicles via deep reinforcement learning. IEEE Syst J. https:\/\/doi.org\/10.1109\/JSYST.2023.3249217","journal-title":"IEEE Syst J"},{"issue":"5","key":"493_CR24","doi-asserted-by":"publisher","first-page":"3017","DOI":"10.1109\/TMC.2021.3125949","volume":"22","author":"J Li","year":"2023","unstructured":"Li J, Liang W, Li Y, Xu Z, Jia X, Guo S (2023) Throughput maximization of delay-aware DNN inference in edge computing by exploring DNN model partitioning and inference parallelism. IEEE Trans Mob Comput 22(5):3017\u20133030","journal-title":"IEEE Trans Mob Comput"},{"key":"493_CR25","doi-asserted-by":"crossref","unstructured":"Liu G, Dai F, Huang B, Qiang Z, Wang S, Li L (2022) A collaborative computation and dependency-aware task offloading method for vehicular edge computing: a reinforcement learning approach. J Cloud Comput 11","DOI":"10.1186\/s13677-022-00340-3"},{"key":"493_CR26","first-page":"1","volume":"2020","author":"J Zhang","year":"2020","unstructured":"Zhang J, Ma B, Huang J (2020) Deploying GIS services into the edge: A study from performance evaluation and optimization viewpoint. Secur Commun Netw 2020:1\u201313","journal-title":"Secur Commun Netw"},{"key":"493_CR27","doi-asserted-by":"publisher","unstructured":"Chen Y, Gu W, Xu J, et al (2022) Dynamic task offloading for digital twin-empowered mobile edge computing via deep reinforcement learning. China Commun. https:\/\/doi.org\/10.23919\/JCC.ea.2022-0372.202302","DOI":"10.23919\/JCC.ea.2022-0372.202302"},{"key":"493_CR28","doi-asserted-by":"crossref","unstructured":"Huang J, Gao H, Wan S, et al (2023) Aoi-aware energy control and computation offloading for industrial IoT. Futur Gener Comput Syst 139:29\u201337","DOI":"10.1016\/j.future.2022.09.007"},{"key":"493_CR29","doi-asserted-by":"crossref","unstructured":"Li S, Zhang N, Jiang R, Zhou Z, Zheng F, Yang G (2022) Joint task offloading and resource allocation in mobile edge computing with energy harvesting. J Cloud Comput Adv Syst Appl 11(1):1\u201314","DOI":"10.1186\/s13677-022-00290-w"},{"key":"493_CR30","doi-asserted-by":"publisher","unstructured":"Zhou Y, Ge H, Ma B et al (2022) Collaborative task offloading and resource allocation with hybrid energy supply for UAV-assisted multi-clouds. J Cloud Comput 11. https:\/\/doi.org\/10.1186\/s13677-022-00317-2","DOI":"10.1186\/s13677-022-00317-2"},{"key":"493_CR31","doi-asserted-by":"publisher","unstructured":"Chen Y, Hu J, Zhao J, Min G (2023) Qos-aware computation offloading in LEO satellite edge computing for IoT: A game-theoretical approach. Chin J Electron. https:\/\/doi.org\/10.23919\/cje.2022.00.412","DOI":"10.23919\/cje.2022.00.412"},{"issue":"10","key":"493_CR32","doi-asserted-by":"publisher","first-page":"4642","DOI":"10.1109\/TII.2018.2843365","volume":"14","author":"G Zhang","year":"2018","unstructured":"Zhang G, Zhang W, Cao Y, Li D, Wang L (2018) Energy-delay tradeoff for dynamic offloading in mobile-edge computing system with energy harvesting devices. IEEE Trans Ind Inform 14(10):4642\u20134655","journal-title":"IEEE Trans Ind Inform"},{"key":"493_CR33","doi-asserted-by":"publisher","unstructured":"Xu J, Chen L, Zhou P (2018) Joint service caching and task offloading for mobile edge computing in dense networks. IEEE INFOCOM 2018 - IEEE Conference on Computer Communications. p 207\u2013215. https:\/\/doi.org\/10.1109\/INFOCOM.2018.8485977","DOI":"10.1109\/INFOCOM.2018.8485977"},{"issue":"4","key":"493_CR34","doi-asserted-by":"publisher","first-page":"1619","DOI":"10.1109\/TNET.2018.2841758","volume":"26","author":"L Chen","year":"2018","unstructured":"Chen L, Zhou S, Xu J (2018) Computation peer offloading for energy-constrained mobile edge computing in small-cell networks. IEEE\/ACM Trans Networking 26(4):1619\u20131632","journal-title":"IEEE\/ACM Trans Networking"},{"issue":"4","key":"493_CR35","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1109\/TPDS.2020.3032443","volume":"32","author":"Z Xu","year":"2021","unstructured":"Xu Z, Zhao L, Liang W, Rana OF, Zhou P, Xia Q, Xu W, Wu G (2021) Energy-aware inference offloading for DNN-driven applications in mobile edge clouds. IEEE Trans Parallel Distrib Syst 32(4):799\u2013814","journal-title":"IEEE Trans Parallel Distrib Syst"},{"issue":"4","key":"493_CR36","doi-asserted-by":"publisher","first-page":"7011","DOI":"10.1109\/JIOT.2019.2913162","volume":"6","author":"J Chen","year":"2019","unstructured":"Chen J, Chen S, Wang Q, Cao B, Feng G, Hu J (2019) iraf: A deep reinforcement learning approach for collaborative mobile edge computing IoT networks. IEEE Internet Things J 6(4):7011\u20137024","journal-title":"IEEE Internet Things J"},{"key":"493_CR37","doi-asserted-by":"publisher","unstructured":"Jeong HJ, Lee HJ, Shin CH, Moon SM (2018) Ionn: Incremental offloading of neural network computations from mobile devices to edge servers. In Proceedings of the ACM Symposium on Cloud Computing (SoCC'18). Association for Computing Machinery, New York. p 401\u2013411. https:\/\/doi.org\/10.1145\/3267809.3267828","DOI":"10.1145\/3267809.3267828"},{"key":"493_CR38","doi-asserted-by":"publisher","unstructured":"Yang Q, Luo X, Li P, Miyazaki T, Wang X (2019) Computation offloading for fast CNN inference in edge computing. In Proceedings of the Conference on Research in Adaptive and Convergent Systems (RACS'19). Association for Computing Machinery, New York. p 101\u2013106. https:\/\/doi.org\/10.1145\/3338840.3355669","DOI":"10.1145\/3338840.3355669"},{"issue":"1","key":"493_CR39","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1145\/584091.584093","volume":"5","author":"CE Shannon","year":"2001","unstructured":"Shannon CE (2001) A mathematical theory of communication. ACM SIGMOBILE Mob Comput Commun Rev 5(1):3\u201355","journal-title":"ACM SIGMOBILE Mob Comput Commun Rev"},{"key":"493_CR40","unstructured":"Krizhevsky A, Hinton G, Sutskever I (2009) Learning multiple layers of features from tiny images. Tech. Rep. TR2009-08, Computer Science Department, University of Toronto"}],"container-title":["Journal of Cloud Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13677-023-00493-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13677-023-00493-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13677-023-00493-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,17]],"date-time":"2023-11-17T18:39:55Z","timestamp":1700246395000},"score":1,"resource":{"primary":{"URL":"https:\/\/journalofcloudcomputing.springeropen.com\/articles\/10.1186\/s13677-023-00493-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,3]]},"references-count":40,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["493"],"URL":"https:\/\/doi.org\/10.1186\/s13677-023-00493-9","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-2901233\/v1","asserted-by":"object"}]},"ISSN":["2192-113X"],"issn-type":[{"value":"2192-113X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,3]]},"assertion":[{"value":"6 May 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 July 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 August 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"116"}}