{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T05:22:59Z","timestamp":1782278579825,"version":"3.54.5"},"reference-count":60,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"3","license":[{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62372487"],"award-info":[{"award-number":["62372487"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Guangdong Natural Science Foundation General Project","award":["2024A1515010378"],"award-info":[{"award-number":["2024A1515010378"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Parallel Distrib. Syst."],"published-print":{"date-parts":[[2025,3]]},"DOI":"10.1109\/tpds.2024.3522333","type":"journal-article","created":{"date-parts":[[2024,12,25]],"date-time":"2024-12-25T19:39:04Z","timestamp":1735155544000},"page":"391-406","source":"Crossref","is-referenced-by-count":4,"title":["Integrated and Fungible Scheduling of Deep Learning Workloads Using Multi-Agent Reinforcement Learning"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0941-9820","authenticated-orcid":false,"given":"Jialun","family":"Li","sequence":"first","affiliation":[{"name":"School of Computer Science, Guangdong Polytechnic Normal University, Guangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6798-9683","authenticated-orcid":false,"given":"Danyang","family":"Xiao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-8640-3530","authenticated-orcid":false,"given":"Diying","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6858-9307","authenticated-orcid":false,"given":"Xuan","family":"Mo","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4714-7021","authenticated-orcid":false,"given":"Weigang","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3234150"},{"key":"ref2","first-page":"947","article-title":"Analysis of large-scale multi-tenant GPU clusters for DNN training workloads","volume-title":"Proc. USENIX Conf. Usenix Annu. Tech. Conf.","author":"Jeon"},{"key":"ref3","first-page":"945","article-title":"MLaaS in the wild: Workload analysis and scheduling in large-scale heterogeneous GPU clusters","volume-title":"Proc. USENIX Symp. Netw. Syst. Des. Implementation","author":"Weng"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3190508.3190517"},{"key":"ref5","first-page":"595","article-title":"Gandiva: Introspective cluster scheduling for deep learning","volume-title":"Proc. USENIX Conf. Operating Syst. Des. Implementation","author":"Xiao"},{"key":"ref6","first-page":"485","article-title":"Tiresias: A GPU cluster manager for distributed deep learning","volume-title":"Proc. USENIX Symp. Netw. Syst. Des. Implementation","author":"Gu"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3342195.3387555"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3342195.3387547"},{"key":"ref9","first-page":"289","article-title":"THEMIS: Fair and efficient GPU cluster scheduling","volume-title":"Proc. USENIX Symp. Netw. Syst. Des. Implementation","author":"Mahajan"},{"key":"ref10","first-page":"481","article-title":"Heterogeneity-aware cluster scheduling policies for deep learning workloads","volume-title":"Proc. USENIX Conf. Operating Syst. Des. Implementation","author":"Narayanan"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.14778\/3446095.3446100"},{"key":"ref12","first-page":"579","article-title":"Looking beyond GPUs for DNN scheduling on multi-tenant clusters","volume-title":"Proc. USENIX Conf. Operating Syst. Des. Implementation","author":"Mohan"},{"key":"ref13","first-page":"1106","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Krizhevsky"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952190"},{"key":"ref18","article-title":"Deep speech: Scaling up end-to-end speech recognition","author":"Hannun","year":"2014"},{"key":"ref19","article-title":"Google\u2019s neural machine translation system: Bridging the gap between human and machine translation","author":"Wu","year":"2016"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D15-1199"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1285"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737460"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3052895"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TNSM.2021.3139607"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3079202"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3138825"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2023.104776"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3207346"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-021-09996-w"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00072"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2023.3317388"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2023.05.032"},{"key":"ref35","first-page":"721","article-title":"Elastic resource sharing for distributed deep learning","volume-title":"Proc. USENIX Symp. Netw. Syst. Des. Implementation","author":"Hwang"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2023.3293835"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3064966"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3104242"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3552326.3587445"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM53939.2023.10229031"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2023.3313779"},{"key":"ref42","first-page":"1008","article-title":"Actor-critic algorithms","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Konda"},{"key":"ref43","first-page":"2790","article-title":"Control of memory, active perception, and action in minecraft","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Oh"},{"key":"ref44","first-page":"1057","article-title":"Policy gradient methods for reinforcement learning with function approximation","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Sutton"},{"key":"ref45","first-page":"1856","article-title":"Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Haarnoja"},{"key":"ref46","first-page":"1352","article-title":"Reinforcement learning with deep energy-based policies","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Haarnoja"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref48","first-page":"764","article-title":"Deep implicit coordination graphs for multi-agent reinforcement learning","volume-title":"Proc. 20th Int. Conf. Auton. Agents Multiagent Syst.","author":"Li"},{"key":"ref49","first-page":"1","article-title":"Semi-supervised classification with graph convolutional networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Kipf"},{"key":"ref50","article-title":"OpenAI Gym","author":"Brockman","year":"2016"},{"key":"ref51","first-page":"1","article-title":"Connection scheduling in web servers","volume-title":"Proc. 2nd USENIX Symp. Internet Technol. Syst.","author":"Crovella"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/2619239.2626334"},{"key":"ref53","first-page":"323","article-title":"Dominant resource fairness: Fair allocation of multiple resource types","volume-title":"Proc. USENIX Symp. Netw. Syst. Des. Implementation","author":"Ghodsi"},{"key":"ref54","first-page":"2961","article-title":"Actor-attention-critic for multi-agent reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Iqbal"},{"key":"ref55","article-title":"Deep learning workload scheduling in GPU datacenters: Taxonomy, challenges and vision","author":"Gao","year":"2022"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1145\/2523616.2523633"},{"key":"ref57","first-page":"1","article-title":"Pollux: Co-adaptive cluster scheduling for goodput-optimized deep learning","volume-title":"Proc. USENIX Conf. Operating Syst. Des. Implementation","author":"Qiao"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/TSC.2017.2679738"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2018.11.012"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2023.3235439"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2023.3276759"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/TCC.2023.3319383"}],"container-title":["IEEE Transactions on Parallel and Distributed Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/71\/10850501\/10816206.pdf?arnumber=10816206","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,29]],"date-time":"2025-01-29T19:10:36Z","timestamp":1738177836000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10816206\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3]]},"references-count":60,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/tpds.2024.3522333","relation":{},"ISSN":["1045-9219","1558-2183","2161-9883"],"issn-type":[{"value":"1045-9219","type":"print"},{"value":"1558-2183","type":"electronic"},{"value":"2161-9883","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3]]}}}