{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T18:18:50Z","timestamp":1776881930053,"version":"3.51.2"},"reference-count":45,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2025,3,22]],"date-time":"2025-03-22T00:00:00Z","timestamp":1742601600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100000038","name":"National Sciences and Engineering Research Council of Canada","doi-asserted-by":"crossref","award":["RGPIN-2022-04754"],"award-info":[{"award-number":["RGPIN-2022-04754"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Sen. Netw."],"published-print":{"date-parts":[[2025,3,31]]},"abstract":"<jats:p>The proliferation of computation-intensive applications, such as autonomous driving, has urged mobile devices to alleviate their local computation pressure using external computing resources. As a promising solution, Multi-access Edge Computing tackles this problem by offloading computational tasks from mobile devices to edge servers. However, existing offloading schemes suffer from two fundamental limitations. First, they lack built-in measures to prevent deadline misses. For safety-critical applications, including autonomous driving, a deadline miss could result in catastrophic consequences. Second, existing schemes typically update offloading policies periodically. Namely, a policy based on the current system state is generated for a time window consisting of multiple time slots. Since system states could change from one time slot to the next one, the generated policy might not work well during the entire window. In this article, we propose a novel offloading scheme for safety-critical applications, Constrained Reinforcement Learning-based Offloading (CRLO). With CRLO, a safety layer is added to the learning-based policy generator, which effectively eliminates deadline misses. Furthermore, a long-sequence forecasting model, Informer, is utilized to predict temporally dependent system states, which helps to generate appropriate offloading policies. Our experimental results indicate that CRLO outperforms existing schemes in terms of deadline satisfaction and task completion time.<\/jats:p>","DOI":"10.1145\/3715695","type":"journal-article","created":{"date-parts":[[2025,1,28]],"date-time":"2025-01-28T11:01:47Z","timestamp":1738062107000},"page":"1-37","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Safety-Critical Offloading with Constrained Reinforcement Learning for Multi-access Edge Computing"],"prefix":"10.1145","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2898-5537","authenticated-orcid":false,"given":"Hui","family":"Huang","sequence":"first","affiliation":[{"name":"Faculty of Computer Science, Dalhousie University, Halifax, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6711-7818","authenticated-orcid":false,"given":"Qiang","family":"Ye","sequence":"additional","affiliation":[{"name":"Faculty of Computer Science, Dalhousie University, Halifax, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8081-1705","authenticated-orcid":false,"given":"Yitong","family":"Zhou","sequence":"additional","affiliation":[{"name":"Faculty of Computer Science, Dalhousie University, Halifax, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,3,22]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCCN.2021.3066619"},{"key":"e_1_3_1_3_2","first-page":"243","volume-title":"Proceedings of the International Conference on Machine Learning (ICML\u201921)","author":"Amani Sanae","year":"2021","unstructured":"Sanae Amani, Christos Thrampoulidis, and Lin Yang. 2021. Safe reinforcement learning with linear function approximation. In Proceedings of the International Conference on Machine Learning (ICML\u201921). 243\u2013253."},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNET.2020.2983119"},{"key":"e_1_3_1_5_2","unstructured":"Petros Christodoulou. 2019. Soft actor-critic for discrete action settings. Retrieved from arXiv:arXiv:1910.07207"},{"key":"e_1_3_1_6_2","first-page":"1","volume-title":"Proceedings of the IEEE International Conference on Computer Communications (INFOCOM\u201921)","author":"Dai Penglin","year":"2021","unstructured":"Penglin Dai, Kaiwen Hu, Xiao Wu, Huanlai Xing, and Zhaofei Yu. 2021. Asynchronous deep reinforcement learning for data-driven task offloading in MEC-empowered vehicular networks. In Proceedings of the IEEE International Conference on Computer Communications (INFOCOM\u201921). 1\u201310."},{"issue":"10","key":"e_1_3_1_7_2","doi-asserted-by":"crossref","first-page":"12175","DOI":"10.1109\/TVT.2020.3013990","article-title":"Edge intelligence for energy-efficient computation offloading and resource allocation in 5G beyond","volume":"69","author":"Dai Yueyue","year":"2020","unstructured":"Yueyue Dai, Ke Zhang, Sabita Maharjan, and Yan. Zhang. 2020. Edge intelligence for energy-efficient computation offloading and resource allocation in 5G beyond. IEEE Trans. Vehic. Technol. 69, 10 (Oct.2020), 12175\u201312186.","journal-title":"IEEE Trans. Vehic. Technol."},{"key":"e_1_3_1_8_2","unstructured":"Gal Dalal Krishnamurthy Dvijotham Matej Vecerik Todd Hester Cosmin Paduraru and Yuval Tassa. 2018. Safe exploration in continuous action spaces. Retrieved from arXiv:https:\/\/arXiv:1801.08757"},{"key":"e_1_3_1_9_2","first-page":"1587","volume-title":"Proceedings of the International Conference on Machine Learning (ICML\u201918)","author":"Fujimoto Scott","year":"2018","unstructured":"Scott Fujimoto, Herke Van Hoof, and David Merger. 2018. Addressing function approximation error in actor-critic methods. In Proceedings of the International Conference on Machine Learning (ICML\u201918). 1587\u20131596."},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2022.3141080"},{"key":"e_1_3_1_11_2","first-page":"4","article-title":"Multi-access edge computing: An overview of ETSI MEC ISG","volume":"1","author":"Giust Fabio","year":"2017","unstructured":"Fabio Giust, Xavier Costa-Perez, and Alex Reznik. 2017. Multi-access edge computing: An overview of ETSI MEC ISG. IEEE 5G Tech Focus 1 (Dec.2017), 4.","journal-title":"IEEE 5G Tech Focus"},{"key":"e_1_3_1_12_2","first-page":"1","volume-title":"Proceedings of the IEEE Global Communications Conference (GLOBECOM\u201918)","author":"Gu Bo","year":"2018","unstructured":"Bo Gu, Zhenyu Zhou, Shahid Mumtaz, Valerio Frascolla, and Ali Kashif Bashir. 2018. Context-aware task offloading for multi-access edge computing: Matching with externalities. In Proceedings of the IEEE Global Communications Conference (GLOBECOM\u201918). 1\u20136."},{"key":"e_1_3_1_13_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2018.2831230"},{"issue":"6","key":"e_1_3_1_14_2","doi-asserted-by":"crossref","first-page":"3870","DOI":"10.1109\/TNSE.2021.3115054","article-title":"Deadline-aware task offloading with partially-observable deep reinforcement learning for multi-access edge computing","volume":"9","author":"Huang Hui","year":"2021","unstructured":"Hui Huang, Qiang Ye, and Yitong. Zhou. 2021. Deadline-aware task offloading with partially-observable deep reinforcement learning for multi-access edge computing. IEEE Trans. Netw. Sci. Eng. 9, 6 (Sep.2021), 3870\u20133885.","journal-title":"IEEE Trans. Netw. Sci. Eng."},{"issue":"3","key":"e_1_3_1_15_2","doi-asserted-by":"crossref","first-page":"1311","DOI":"10.1109\/TNSE.2022.3188921","article-title":"6G-Empowered offloading for realtime applications in multi-access edge computing","volume":"10","author":"Huang Hui","year":"2022","unstructured":"Hui Huang, Qiang Ye, and Yitong Zhou. 2022. 6G-Empowered offloading for realtime applications in multi-access edge computing. IEEE Trans. Netw. Sci. Eng. 10, 3 (July2022), 1311\u20131325.","journal-title":"IEEE Trans. Netw. Sci. Eng."},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/JSYST.2015.2446205"},{"issue":"10","key":"e_1_3_1_17_2","doi-asserted-by":"crossref","first-page":"7145","DOI":"10.1109\/TII.2021.3052531","article-title":"A survey of computational intelligence for 6G: Key technologies, applications and trends","volume":"17","author":"Ji Baofeng","year":"2021","unstructured":"Baofeng Ji, Yanan Wang, Kang Song, Chunguo Li, Hong Wen, Varun G. Menon, and Shahid Mumtaz. 2021. A survey of computational intelligence for 6G: Key technologies, applications and trends. IEEE Trans. Industr. Inform. 17, 10 (Jan.2021), 7145\u20137154.","journal-title":"IEEE Trans. Industr. Inform."},{"issue":"17","key":"e_1_3_1_18_2","first-page":"4000","article-title":"Joint task offloading and resource allocation for energy-constrained mobile edge computing","volume":"22","author":"Jiang Hongbo","year":"2022","unstructured":"Hongbo Jiang, Xingxia. Dai, Zhu Xiao, and Arun Iyengar. 2022. Joint task offloading and resource allocation for energy-constrained mobile edge computing. IEEE Trans. Mobile Comput. 22, 17 (Feb.2022), 4000\u20134015.","journal-title":"IEEE Trans. Mobile Comput."},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2018.2880250"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNET.2020.2968209"},{"issue":"9","key":"e_1_3_1_21_2","doi-asserted-by":"crossref","first-page":"6308","DOI":"10.1109\/TII.2022.3155162","article-title":"Deep reinforcement learning based energy efficient edge computing for Internet of Vehicles","volume":"18","author":"Kong Xiangjie","year":"2022","unstructured":"Xiangjie Kong, Gaohui Duan, Mingliang Hou, Guojiang Shen, Hui Wang, Xiaoran Yan, and Mario Collotta. 2022. Deep reinforcement learning based energy efficient edge computing for Internet of Vehicles. IEEE Trans. Industr. Inform. 18, 9 (Mar.2022), 6308\u20136316.","journal-title":"IEEE Trans. Industr. Inform."},{"issue":"1","key":"e_1_3_1_22_2","first-page":"278","article-title":"QoS driven task offloading with statistical guarantee in mobile edge computing","volume":"21","author":"Li Qing","year":"2022","unstructured":"Qing Li, Shangguang Wang, Ao Zhang, Xiao. Ma, Fangchun Yang, and Alex X. Liu. 2022. QoS driven task offloading with statistical guarantee in mobile edge computing. IEEE Trans. Mobile Comput. 21, 1 (June2022), 278\u2013290.","journal-title":"IEEE Trans. Mobile Comput."},{"key":"e_1_3_1_23_2","first-page":"1","volume-title":"Proceedings of International Conference Learning Representations (ICLR\u201916)","author":"Lillicrap Timothy P.","year":"2016","unstructured":"Timothy P. Lillicrap, Jonathan J. Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez Yuval Tassa, David Silver, and Daan Wierstra. 2016. Continuous control with deep reinforcement learning. In Proceedings of International Conference Learning Representations (ICLR\u201916). 1\u201314."},{"issue":"4","key":"e_1_3_1_24_2","first-page":"2147","article-title":"Efficient dependent task offloading for multiple applications in MEC-cloud system","volume":"22","author":"Liu Jiagang","year":"2021","unstructured":"Jiagang Liu, Ju Ren, Yongmin Zhang, Xuhong Peng, Yaoxue Zhang, and Yuanyuan Yang. 2021. Efficient dependent task offloading for multiple applications in MEC-cloud system. IEEE Trans. Mobile Comput. 22, 4 (Oct.2021), 2147\u20132162.","journal-title":"IEEE Trans. Mobile Comput."},{"issue":"2","key":"e_1_3_1_25_2","first-page":"2169","article-title":"Mobility-aware multi-hop task offloading for autonomous driving in vehicular edge computing and networks","volume":"24","author":"Liu Lei","year":"2022","unstructured":"Lei Liu, Ming Zhao, Miao Yu, Mian Ahmad Jan, Dapeng Lan, and Amirhosein Taherkordi. 2022. Mobility-aware multi-hop task offloading for autonomous driving in vehicular edge computing and networks. IEEE Trans. Intell. Transport. Syst. 24, 2 (Jan.2022), 2169\u20132182.","journal-title":"IEEE Trans. Intell. Transport. Syst."},{"key":"e_1_3_1_26_2","first-page":"13644","volume-title":"Proceedings of IEEE Conference on Machine Learning (ICML\u201922)","author":"Liu Zuxin","year":"2022","unstructured":"Zuxin Liu, Zhepeng Cen, Vladislav Isenbaev, Wei Liu, Steven Wu, Bo Li, and Ding Zhao. 2022. Constrained variational policy optimization for safe reinforcement learning. In Proceedings of IEEE Conference on Machine Learning (ICML\u201922). 13644\u201313668."},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2021.3106401"},{"key":"e_1_3_1_28_2","first-page":"342","article-title":"Dynamic computation offloading in multi-access edge computing via ultra-reliable and low-latency communications","volume":"6","author":"Merluzzi Mattia","year":"2020","unstructured":"Mattia Merluzzi, Lorenzo Paolo Di, Sergio Barbarossa, and Valerio Frascolla. 2020. Dynamic computation offloading in multi-access edge computing via ultra-reliable and low-latency communications. IEEE Trans. Signal Inf. Process. Netw. 6 (Mar.2020), 342\u2013356.","journal-title":"IEEE Trans. Signal Inf. Process. Netw."},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2019.2908403"},{"issue":"8","key":"e_1_3_1_30_2","doi-asserted-by":"crossref","first-page":"6611","DOI":"10.1109\/JIOT.2022.3153399","article-title":"Energy efficient computation offloading with DVFS using deep reinforcement learning for time-critical IoT applications in edge computing","volume":"10","author":"Panda Saroj Kumar","year":"2022","unstructured":"Saroj Kumar Panda, Man Lin, and Ti Zhou. 2022. Energy efficient computation offloading with DVFS using deep reinforcement learning for time-critical IoT applications in edge computing. IEEE Internet Things J. 10, 8 (Feb.2022), 6611\u20136621.","journal-title":"IEEE Internet Things J."},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVT.2019.2924015"},{"issue":"3","key":"e_1_3_1_32_2","doi-asserted-by":"crossref","first-page":"1414","DOI":"10.1109\/TETC.2021.3090061","article-title":"Offloading decision for mobile multi-access edge computing in a multi-tiered 6G network","volume":"10","author":"Rodrigues Tiago Koketsu","year":"2021","unstructured":"Tiago Koketsu Rodrigues, Jiajia Liu, and Nei Kato. 2021. Offloading decision for mobile multi-access edge computing in a multi-tiered 6G network. IEEE Trans. Emerg. Top. Comput. 10, 3 (June2021), 1414\u20131427.","journal-title":"IEEE Trans. Emerg. Top. Comput."},{"key":"e_1_3_1_33_2","first-page":"37","volume-title":"Proceedings of the IEEE Conference on Computer Communications (INFOCOM\u201918)","author":"Sundar Sowndarya","year":"2018","unstructured":"Sowndarya Sundar and Ben Liang. 2018. Offloading dependent tasks with communication delay and deadline constraint. In Proceedings of the IEEE Conference on Computer Communications (INFOCOM\u201918). 37\u201345."},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2020.3036871"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/TWC.2017.2785305"},{"key":"e_1_3_1_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2020.3014896"},{"key":"e_1_3_1_37_2","article-title":"Parameterized deep reinforcement learning with hybrid action space for edge task offloading","author":"Wang Ting","year":"2023","unstructured":"Ting Wang, Yuxiang Deng, Zhao Yang, Yang Wang, and Haibin Cai. 2023. Parameterized deep reinforcement learning with hybrid action space for edge task offloading. IEEE Internet Things J. (Mar.2023). DOI:10.1109\/JIOT.2023.3327121.","journal-title":"IEEE Internet Things J."},{"key":"e_1_3_1_38_2","first-page":"22419","volume-title":"Proceedings of the Advances in Neural Information Processing Systems","volume":"34","author":"Wu Haixu","year":"2021","unstructured":"Haixu Wu, Jiehui Xu, Jianmin Wang, and Mingsheng Long. 2021. Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. In Proceedings of the Advances in Neural Information Processing Systems, Vol. 34. 22419\u201322430."},{"issue":"3","key":"e_1_3_1_39_2","first-page":"1256","article-title":"AI-driven and MEC-empowered confident information coverage hole recovery in 6G-enabled IoT","volume":"10","author":"Xia Yunzhi","year":"2022","unstructured":"Yunzhi Xia, Xianjun Deng, Lingzhi Yi, Laurence T. Yang, Tang Xiao, Chelu Zhu, and Zhongping Tian. 2022. AI-driven and MEC-empowered confident information coverage hole recovery in 6G-enabled IoT. IEEE Trans. Netw. Sci. Eng. 10, 3 (Feb.2022), 1256\u20131269.","journal-title":"IEEE Trans. Netw. Sci. Eng."},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2021.05.017"},{"issue":"9","key":"e_1_3_1_41_2","doi-asserted-by":"crossref","first-page":"2745","DOI":"10.1109\/TMC.2020.2990630","article-title":"Computation offloading in multi-access edge computing: A multi-task learning approach","volume":"20","author":"Yang Bo","year":"2021","unstructured":"Bo Yang, Xuelin Cao, Joshua Bassey, Xiangfang Li, and Lijun Qian. 2021. Computation offloading in multi-access edge computing: A multi-task learning approach. IEEE Trans. Mobile Comput. 20, 9 (Apr.2021), 2745\u20132762.","journal-title":"IEEE Trans. Mobile Comput."},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.2969148"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSC.2018.2867482"},{"key":"e_1_3_1_44_2","unstructured":"Jinkai Zheng Tom H. Luan Longxiang Gao Yao Zhang and Yuan Wu. 2021. Learning based task offloading in digital twin empowered internet of vehicles. Retrieved from arXiv:https:\/\/arXiv:2201.09076"},{"key":"e_1_3_1_45_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCOMM.2014.2357423"},{"key":"e_1_3_1_46_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i12.17325"}],"container-title":["ACM Transactions on Sensor Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3715695","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3715695","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T01:19:14Z","timestamp":1750295954000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3715695"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,22]]},"references-count":45,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,3,31]]}},"alternative-id":["10.1145\/3715695"],"URL":"https:\/\/doi.org\/10.1145\/3715695","relation":{},"ISSN":["1550-4859","1550-4867"],"issn-type":[{"value":"1550-4859","type":"print"},{"value":"1550-4867","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,22]]},"assertion":[{"value":"2023-06-05","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-11-19","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-22","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}