Accuracy Certificates for Convex Minimization with Inexact Oracle.
Saved in:
| Title: | Accuracy Certificates for Convex Minimization with Inexact Oracle. |
|---|---|
| Authors: | Gladin, Egor1,2 (AUTHOR) elgladin@hse.ru, Gasnikov, Alexander3,4,5 (AUTHOR) gasnikov@yandex.ru, Dvurechensky, Pavel6 (AUTHOR) pavel.dvurechensky@wias-berlin.de |
| Source: | Journal of Optimization Theory & Applications. Jan2025, Vol. 204 Issue 1, p1-23. 23p. |
| Subjects: | Lagrange problem, Subgradient methods, Algorithms |
| Abstract: | Accuracy certificates for convex minimization problems allow for online verification of the accuracy of approximate solutions and provide a theoretically valid online stopping criterion. When solving the Lagrange dual problem, accuracy certificates produce a simple way to recover an approximate primal solution and estimate its accuracy. In this paper, we generalize accuracy certificates for the setting of an inexact first-order oracle, including the setting of primal and Lagrange dual pair of problems. We further propose an explicit way to construct accuracy certificates for a large class of cutting plane methods based on polytopes. As a by-product, we show that the considered cutting plane methods can be efficiently used with a noisy oracle even though they were originally designed to be equipped with an exact oracle. Finally, we illustrate the work of the proposed certificates in the numerical experiments highlighting that our certificates provide a tight upper bound on the objective residual. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Optimization Theory & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 181861227 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Accuracy Certificates for Convex Minimization with Inexact Oracle. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Gladin%2C+Egor%22">Gladin, Egor</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> elgladin@hse.ru</i><br /><searchLink fieldCode="AR" term="%22Gasnikov%2C+Alexander%22">Gasnikov, Alexander</searchLink><relatesTo>3,4,5</relatesTo> (AUTHOR)<i> gasnikov@yandex.ru</i><br /><searchLink fieldCode="AR" term="%22Dvurechensky%2C+Pavel%22">Dvurechensky, Pavel</searchLink><relatesTo>6</relatesTo> (AUTHOR)<i> pavel.dvurechensky@wias-berlin.de</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Journal+of+Optimization+Theory+%26+Applications%22">Journal of Optimization Theory & Applications</searchLink>. Jan2025, Vol. 204 Issue 1, p1-23. 23p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Lagrange+problem%22">Lagrange problem</searchLink><br /><searchLink fieldCode="DE" term="%22Subgradient+methods%22">Subgradient methods</searchLink><br /><searchLink fieldCode="DE" term="%22Algorithms%22">Algorithms</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Accuracy certificates for convex minimization problems allow for online verification of the accuracy of approximate solutions and provide a theoretically valid online stopping criterion. When solving the Lagrange dual problem, accuracy certificates produce a simple way to recover an approximate primal solution and estimate its accuracy. In this paper, we generalize accuracy certificates for the setting of an inexact first-order oracle, including the setting of primal and Lagrange dual pair of problems. We further propose an explicit way to construct accuracy certificates for a large class of cutting plane methods based on polytopes. As a by-product, we show that the considered cutting plane methods can be efficiently used with a noisy oracle even though they were originally designed to be equipped with an exact oracle. Finally, we illustrate the work of the proposed certificates in the numerical experiments highlighting that our certificates provide a tight upper bound on the objective residual. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Journal of Optimization Theory & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=181861227 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s10957-024-02599-9 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 23 StartPage: 1 Subjects: – SubjectFull: Lagrange problem Type: general – SubjectFull: Subgradient methods Type: general – SubjectFull: Algorithms Type: general Titles: – TitleFull: Accuracy Certificates for Convex Minimization with Inexact Oracle. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Gladin, Egor – PersonEntity: Name: NameFull: Gasnikov, Alexander – PersonEntity: Name: NameFull: Dvurechensky, Pavel IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Text: Jan2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 00223239 Numbering: – Type: volume Value: 204 – Type: issue Value: 1 Titles: – TitleFull: Journal of Optimization Theory & Applications Type: main |
| ResultId | 1 |