Publications
Detailed Information
Plug-and-play methods provably converge with properly trained denoisers
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Ryu, Ernest K. | - |
dc.contributor.author | Liu, Jialin | - |
dc.contributor.author | Wang, Sicheng | - |
dc.contributor.author | Chen, Xiaohan | - |
dc.contributor.author | Wang, Zhangyang | - |
dc.contributor.author | Yin, Wotao | - |
dc.date.accessioned | 2023-07-14T04:16:31Z | - |
dc.date.available | 2023-07-14T04:16:31Z | - |
dc.date.created | 2023-07-13 | - |
dc.date.issued | 2019-01 | - |
dc.identifier.citation | 36th International Conference on Machine Learning, ICML 2019, Vol.2019-June, pp.9764-9775 | - |
dc.identifier.uri | https://hdl.handle.net/10371/195140 | - |
dc.description.abstract | © 2019 International Machine Learning Society (IMLS).Plug-and-play (PnP) is a non-convex framework that integrates modern denoising priors, such as BM3D or deep learning-based denoisers, into ADMM or other proximal algorithms. An advantage of PnP is that one can use pre-trained denoisers when there is not sufficient data for end-to-end training. Although PnP has been recently studied extensively with great empirical success, theoretical analysis addressing even the most basic question of convergence has been insufficient. In this paper, we theoretically establish convergence of PnP-FBS and PnP-ADMM, without using diminishing stepsizes, under a certain Lipschitz condition on the denoisers. We then propose real spectral normalization, a technique for training deep learning-based denoisers to satisfy the proposed Lipschitz condition. Finally, we present experimental results validating the theory. | - |
dc.language | 영어 | - |
dc.publisher | International Machine Learning Society (IMLS) | - |
dc.title | Plug-and-play methods provably converge with properly trained denoisers | - |
dc.type | Article | - |
dc.citation.journaltitle | 36th International Conference on Machine Learning, ICML 2019 | - |
dc.identifier.scopusid | 2-s2.0-85075748815 | - |
dc.citation.endpage | 9775 | - |
dc.citation.startpage | 9764 | - |
dc.citation.volume | 2019-June | - |
dc.description.isOpenAccess | N | - |
dc.contributor.affiliatedAuthor | Ryu, Ernest K. | - |
dc.description.journalClass | 1 | - |
- Appears in Collections:
- Files in This Item:
- There are no files associated with this item.
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.