{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,19]],"date-time":"2026-01-19T03:15:33Z","timestamp":1768792533350,"version":"3.49.0"},"reference-count":47,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"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":["62025208"],"award-info":[{"award-number":["62025208"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62476157"],"award-info":[{"award-number":["62476157"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62472302"],"award-info":[{"award-number":["62472302"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Special Project for Guiding the Transformation of Scientific and Technological Achievements of Shanxi Province","award":["202304021301037"],"award-info":[{"award-number":["202304021301037"]}]},{"name":"Central Government Guiding Local Scientific and Technological Development Fund of Shanxi Province","award":["YDZJSX2024C003"],"award-info":[{"award-number":["YDZJSX2024C003"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Knowl. Data Eng."],"published-print":{"date-parts":[[2025,5]]},"DOI":"10.1109\/tkde.2025.3543225","type":"journal-article","created":{"date-parts":[[2025,2,18]],"date-time":"2025-02-18T18:26:02Z","timestamp":1739903162000},"page":"2831-2845","source":"Crossref","is-referenced-by-count":2,"title":["PipeOptim: Ensuring Effective 1F1B Schedule With Optimizer-Dependent Weight Prediction"],"prefix":"10.1109","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9041-3244","authenticated-orcid":false,"given":"Lei","family":"Guan","sequence":"first","affiliation":[{"name":"College of Computer Science and Technology, Taiyuan University of Technology, Taiyuan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9743-2034","authenticated-orcid":false,"given":"Dongsheng","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1000-1109","authenticated-orcid":false,"given":"Yongle","family":"Chen","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Taiyuan University of Technology, Taiyuan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5887-9327","authenticated-orcid":false,"given":"Jiye","family":"Liang","sequence":"additional","affiliation":[{"name":"School of Computer and Information Technology, Shanxi University, Taiyuan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8276-3639","authenticated-orcid":false,"given":"Wenjian","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer and Information Technology, Shanxi University, Taiyuan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xicheng","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Computer, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01212"},{"key":"ref3","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"key":"ref4","article-title":"Language models are few-shot learners","author":"Brown","year":"2020"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2889052"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054345"},{"key":"ref7","article-title":"Accurate, large minibatch SGD: Training imagenet in 1 hour","author":"Goyal","year":"2017"},{"key":"ref8","article-title":"Scaling SGD batch size to 32k for ImageNet training","author":"You","year":"2017"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.48550\/arxiv.1811.06965"},{"key":"ref10","first-page":"1","article-title":"Pipedream: Generalized pipeline parallelism for DNN training","volume-title":"Proc. 27th ACM Symp. Operating Syst. Princ.","author":"Narayanan"},{"key":"ref11","first-page":"7937","article-title":"Memory-efficient pipeline-parallel DNN training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Narayanan"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3387399"},{"key":"ref13","article-title":"Efficient and robust parallel dnn training through model parallelism on multi-GPU platform","author":"Chen","year":"2018"},{"key":"ref14","article-title":"XPipe: Efficient pipeline model parallelism for multi-GPU DNN training","author":"Guan","year":"2019"},{"issue":"1","key":"ref15","doi-asserted-by":"crossref","first-page":"145","DOI":"10.1016\/S0893-6080(98)00116-6","article-title":"On the momentum term in gradient descent learning algorithms","volume":"12","author":"Qian","year":"1999","journal-title":"Neural Netw."},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/s11390-024-3872-3"},{"key":"ref17","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"ref18","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2017"},{"issue":"5","key":"ref19","article-title":"Tiny ImageNet classification with convolutional neural networks","volume":"2","author":"Yao","year":"2015","journal-title":"CS 231N"},{"key":"ref20","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/W16-2323","article-title":"Edinburgh neural machine translation systems for WMT 16","author":"Sennrich","year":"2016"},{"key":"ref21","first-page":"142","article-title":"Learning word vectors for sentiment analysis","volume-title":"Proc. 49th Annu. Meeting Assoc. Comput. Linguistics-Hum. Lang. Technol.","author":"Maas"},{"key":"ref22","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Krizhevsky"},{"key":"ref23","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.4324\/9781410605337-29"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-477"},{"key":"ref28","article-title":"Google\u2019s neural machine translation system: Bridging the gap between human and machine translation","author":"Wu","year":"2016"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TENCON.2019.8929465"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476145"},{"key":"ref32","article-title":"Ampnet: Asynchronous model-parallel training for dynamic neural networks","author":"Gaunt","year":"2017"},{"key":"ref33","article-title":"Demystifying parallel and distributed deep learning: An in-depth concurrency analysis","author":"Ben-Nun","year":"2018"},{"key":"ref34","first-page":"269","article-title":"Pipemare: Asynchronous pipeline parallel DNN training","volume":"3","author":"Yang","year":"2021","journal-title":"Proc. Mach. Learn. Syst."},{"key":"ref35","first-page":"4150","article-title":"Pipetransformer: Automated elastic pipelining for distributed training of large-scale models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"He"},{"key":"ref36","first-page":"431","article-title":"DAPPLE: A pipelined data parallel approach for training large models","volume-title":"Proc. 26th ACM SIGPLAN Symp. Princ. Pract. Parallel Program.","author":"Fan"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00049"},{"key":"ref38","first-page":"1","article-title":"Hanayo: Harnessing wave-like pipeline parallelism for enhanced large model training efficiency","volume-title":"Proc. Int. Conf. High Perform. Comput. Netw. Storage Anal.","author":"Liu"},{"key":"ref39","first-page":"708","article-title":"PipeFisher: Efficient training of large language models using pipelining and fisher information matrices","volume":"5","author":"Osawa","year":"2023","journal-title":"Proc. Mach. Learn. Syst."},{"key":"ref40","first-page":"1","article-title":"Zero bubble (almost) pipeline parallelism","volume-title":"Proc. 12th Int. Conf. Learn. Representations","author":"Qi"},{"key":"ref41","first-page":"1","article-title":"Group-based interleaved pipeline parallelism for large-scale DNN training","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Yang"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3572848.3577484"},{"key":"ref43","first-page":"685","article-title":"Deep learning with elastic averaging SGD","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhang"},{"key":"ref44","first-page":"1223","article-title":"Large scale distributed deep networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Dean"},{"key":"ref45","first-page":"1","article-title":"Efficient large-scale language model training on GPU clusters using megatron-LM","volume-title":"Proc. Int. Conf. High Perform. Comput. Netw. Storage Anal.","author":"Narayanan"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-33374-3_26"},{"key":"ref47","first-page":"381","article-title":"Fine-tuning giant neural networks on commodity hardware with automatic pipeline model parallelism","volume-title":"Proc. 2021 USENIX Annu. Tech. Conf.","author":"Eliad"}],"container-title":["IEEE Transactions on Knowledge and Data Engineering"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/69\/10948402\/10891748.pdf?arnumber=10891748","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,7]],"date-time":"2025-04-07T03:24:09Z","timestamp":1743996249000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10891748\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5]]},"references-count":47,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tkde.2025.3543225","relation":{},"ISSN":["1041-4347","1558-2191","2326-3865"],"issn-type":[{"value":"1041-4347","type":"print"},{"value":"1558-2191","type":"electronic"},{"value":"2326-3865","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5]]}}}