dc.contributor.author | Mai, The Tien | |
dc.date.accessioned | 2023-01-09T07:37:36Z | |
dc.date.available | 2023-01-09T07:37:36Z | |
dc.date.created | 2022-08-05T13:58:47Z | |
dc.date.issued | 2022 | |
dc.identifier.citation | Lecture Notes in Networks and Systems. 2022, 506 545-559. | en_US |
dc.identifier.issn | 2367-3370 | |
dc.identifier.uri | https://hdl.handle.net/11250/3041733 | |
dc.description.abstract | In this paper, we generalize the problem of single-index model to the context of continual learning in which a learner is challenged with a sequence of tasks one by one and the dataset of each task is revealed in an online fashion. We propose a randomized strategy that is able to learn a common single-index (meta-parameter) for all tasks and a specific link function for each task. The common single-index allows to transfer the information gained from the previous tasks to a new one. We provide a rigorous theoretical analysis of our proposed strategy by proving some regret bounds under different assumption on the loss function. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | Springer | en_US |
dc.title | On Regret Bounds for Continual Single-Index Learning | en_US |
dc.title.alternative | On Regret Bounds for Continual Single-Index Learning | en_US |
dc.type | Journal article | en_US |
dc.description.version | submittedVersion | en_US |
dc.source.pagenumber | 545-559 | en_US |
dc.source.volume | 506 | en_US |
dc.source.journal | Lecture Notes in Networks and Systems | en_US |
dc.identifier.doi | 10.1007/978-3-031-10461-9_37 | |
dc.identifier.cristin | 2041382 | |
cristin.ispublished | true | |
cristin.fulltext | preprint | |
cristin.qualitycode | 1 | |