{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:36:23Z","timestamp":1777890983126,"version":"3.51.4"},"reference-count":41,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2022,11,3]],"date-time":"2022-11-03T00:00:00Z","timestamp":1667433600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Big Data"],"abstract":"<jats:p>Dynamic transfer learning refers to the knowledge transfer from a static source task with adequate label information to a dynamic target task with little or no label information. However, most existing theoretical studies and practical algorithms of dynamic transfer learning assume that the target task is continuously evolving over time. This strong assumption is often violated in real world applications, e.g., the target distribution is suddenly changing at some time stamp. To solve this problem, in this paper, we propose a novel meta-learning framework <jats:monospace>L2S<\/jats:monospace> based on a progressive meta-task scheduler for dynamic transfer learning. The crucial idea of <jats:monospace>L2S<\/jats:monospace> is to incrementally learn to schedule the meta-pairs of tasks and then learn the optimal model initialization from those meta-pairs of tasks for fast adaptation to the newest target task. The effectiveness of our <jats:monospace>L2S<\/jats:monospace> framework is verified both theoretically and empirically.<\/jats:p>","DOI":"10.3389\/fdata.2022.1052972","type":"journal-article","created":{"date-parts":[[2022,11,3]],"date-time":"2022-11-03T09:58:06Z","timestamp":1667469486000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Dynamic transfer learning with progressive meta-task scheduler"],"prefix":"10.3389","volume":"5","author":[{"given":"Jun","family":"Wu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingrui","family":"He","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2022,11,3]]},"reference":[{"key":"B1","first-page":"32","article-title":"\u201cMemory efficient online meta learning,\u201d","author":"Acar","year":"2021","journal-title":"International Conference on Machine Learning"},{"key":"B2","first-page":"66","article-title":"\u201cf-domain adversarial learning: theory and algorithms,\u201d","author":"Acuna","year":"2021","journal-title":"International Conference on Machine Learning"},{"key":"B3","doi-asserted-by":"publisher","first-page":"151","DOI":"10.1007\/s10994-009-5152-4","article-title":"A theory of learning from different domains","volume":"79","author":"Ben-David","year":"2010","journal-title":"Mach. Learn"},{"key":"B4","doi-asserted-by":"publisher","first-page":"2128","DOI":"10.1109\/TKDE.2016.2551241","article-title":"Incremental evolving domain adaptation","volume":"28","author":"Bitarafan","year":"2016","journal-title":"IEEE Trans. Knowl. Data Eng"},{"key":"B5","article-title":"\u201cAdapting to continuously shifting domains,\u201d","volume-title":"International Conference on Learning Representations Workshop","author":"Bobu","year":"2018"},{"key":"B6","first-page":"8201","article-title":"\u201cGradual domain adaptation without indexed intermediate domains,\u201d","author":"Chen","year":"2021","journal-title":"Advances in Neural Information Processing Systems, Vol. 34"},{"key":"B7","first-page":"1126","article-title":"\u201cModel-agnostic meta-learning for fast adaptation of deep networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Finn","year":"2017"},{"key":"B8","article-title":"\u201cProbabilistic model-agnostic meta-learning,\u201d","volume-title":"Advances in Neural Information Processing Systems","author":"Finn","year":"2018"},{"key":"B9","first-page":"1920","article-title":"\u201cOnline meta-learning,\u201d","volume-title":"International Conference on Machine Learning","author":"Finn","year":"2019"},{"key":"B10","doi-asserted-by":"publisher","first-page":"2096","DOI":"10.1007\/978-3-319-58347-1_10","article-title":"Domain-adversarial training of neural networks","volume":"17","author":"Ganin","year":"2016","journal-title":"J. Mach. Learn. Res"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1414","DOI":"10.1109\/TPAMI.2016.2599532","article-title":"Scatter component analysis: a unified framework for domain adaptation and domain generalization","volume":"39","author":"Ghifary","year":"2016","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"B12","first-page":"723","article-title":"A kernel two-sample test","volume":"13","author":"Gretton","year":"2012","journal-title":"J. Mach. Learn. Res"},{"key":"B13","first-page":"770","article-title":"\u201cDeep residual learning for image recognition,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"He","year":"2016"},{"key":"B14","first-page":"867","article-title":"\u201cContinuous manifold based adaptation for evolving visual domains,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Hoffman","year":"2014"},{"key":"B15","doi-asserted-by":"publisher","first-page":"5149","DOI":"10.1109\/TPAMI.2021.3079209","article-title":"Meta-learning in neural networks: a survey","volume":"14","author":"Hospedales","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"B16","first-page":"5468","article-title":"\u201cUnderstanding self-training for gradual domain adaptation,\u201d","author":"Kumar","year":"2020","journal-title":"International Conference on Machine Learning"},{"key":"B17","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"B18","first-page":"22338","article-title":"\u201cLearning to adapt to evolving domains,\u201d","author":"Liu","year":"2020","journal-title":"Advances in Neural Information Processing Systems, Vol. 33"},{"key":"B19","first-page":"97","article-title":"\u201cLearning transferable features with deep adaptation networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Long","year":"2015"},{"key":"B20","first-page":"2208","article-title":"\u201cDeep transfer learning with joint adaptation networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Long","year":"2017"},{"key":"B21","doi-asserted-by":"publisher","first-page":"2346","DOI":"10.1109\/TKDE.2018.2876857","article-title":"Learning under concept drift: a review","volume":"31","author":"Lu","year":"2018","journal-title":"IEEE Trans. Knowl. Data Eng"},{"key":"B22","first-page":"6568","article-title":"\u201cAdagraph: unifying predictive and continuous domain adaptation through graphs,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Mancini","year":"2019"},{"key":"B23","article-title":"\u201cDomain adaptation: learning bounds and algorithms,\u201d","volume-title":"22nd Conference on Learning Theory, COLT 2009","author":"Mansour","year":"2009"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2009","journal-title":"IEEE Trans. Knowl. Data Eng"},{"key":"B25","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v32i1.11784","article-title":"\u201cWasserstein distance guided representation learning for domain adaptation,\u201d","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 32","author":"Shen","year":"2018"},{"key":"B26","doi-asserted-by":"publisher","first-page":"2665","DOI":"10.1609\/aaai.v35i3.16370","article-title":"Gradient regularized contrastive learning for continual domain adaptation","volume":"35","author":"Tang","year":"2021","journal-title":"Proc. AAAI Conf. Artif. Intell"},{"key":"B27","first-page":"3740","article-title":"\u201cUnsupervised continual learning for gradually varying domains,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Taufique","year":"2022"},{"key":"B28","first-page":"7852","article-title":"\u201cOn the theory of transfer learning: The importance of task diversity,\u201d","author":"Tripuraneni","year":"2020","journal-title":"Advances in Neural Information Processing Systems, Vol. 33"},{"key":"B29","first-page":"22784","article-title":"\u201cUnderstanding gradual domain adaptation: improved analysis, optimal path and beyond,\u201d","volume-title":"Proceedings of the 39th International Conference on Machine Learning","author":"Wang","year":"2022"},{"key":"B30","first-page":"9898","article-title":"\u201cContinuously indexed domain adaptation,\u201d","author":"Wang","year":"2020","journal-title":"Proceedings of the 37th International Conference on Machine Learning"},{"key":"B31","first-page":"9837","article-title":"\u201cOn the global optimality of model-agnostic meta-learning,\u201d","author":"Wang","year":"2020","journal-title":"International Conference on Machine Learning"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2006.03230","article-title":"Continuous transfer learning with label-informed distribution alignment","author":"Wu","year":"2020","journal-title":"arXiv preprint arXiv:2006.03230"},{"key":"B33","first-page":"2039","article-title":"\u201cDomain adaptation with dynamic open-set targets,\u201d","volume-title":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","author":"Wu","year":""},{"key":"B34","first-page":"3573","article-title":"\u201cA unified meta-learning framework for dynamic transfer learning,\u201d","volume-title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22","author":"Wu","year":""},{"key":"B35","first-page":"1852","article-title":"\u201cIndirect invisible poisoning attacks on domain adaptation,\u201d","author":"Wu","year":"2021","journal-title":"Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining"},{"key":"B36","doi-asserted-by":"crossref","first-page":"4489","DOI":"10.1109\/ICRA.2018.8460982","article-title":"\u201cIncremental adversarial domain adaptation for continually changing environments,\u201d","volume-title":"2018 IEEE International Conference on Robotics and Automation (ICRA)","author":"Wulfmeier","year":"2018"},{"key":"B37","first-page":"7497","article-title":"\u201cMeta-learning with an adaptive task scheduler,\u201d","author":"Yao","year":"2021","journal-title":"Advances in Neural Information Processing Systems, Vol. 34"},{"key":"B38","first-page":"7404","article-title":"\u201cBridging theory and algorithm for domain adaptation,\u201d","volume-title":"International Conference on Machine Learning","author":"Zhang","year":"2019"},{"key":"B39","first-page":"3231","article-title":"\u201cOptimizing the wisdom of the crowd: Inference, learning, and teaching,\u201d","volume-title":"Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","author":"Zhou","year":""},{"key":"B40","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3310227","article-title":"Multi-task crowdsourcing via an optimization framework","volume":"13","author":"Zhou","year":"","journal-title":"ACM Trans. Knowl. Discov. Data"},{"key":"B41","first-page":"579","article-title":"\u201cMultiCmbox 2: An optimization framework for learning from task and worker dual heterogeneity,\u201d","volume-title":"Proceedings of the 2017 SIAM International Conference on Data Mining","author":"Zhou","year":"2017"}],"container-title":["Frontiers in Big Data"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2022.1052972\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,11,3]],"date-time":"2022-11-03T09:59:03Z","timestamp":1667469543000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2022.1052972\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,3]]},"references-count":41,"alternative-id":["10.3389\/fdata.2022.1052972"],"URL":"https:\/\/doi.org\/10.3389\/fdata.2022.1052972","relation":{},"ISSN":["2624-909X"],"issn-type":[{"value":"2624-909X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,11,3]]},"article-number":"1052972"}}