ℓp-norm support vector data description

Simple item page
buir.contributor.authorArashloo, Shervin Rahimzadeh
buir.contributor.orcidArashloo, Shervin Rahimzadeh|0000-0003-0189-4774
dc.citation.epage108930- 11en_US
dc.citation.spage108930- 1en_US
dc.citation.volumeNumber132en_US
dc.contributor.authorArashloo, Shervin Rahimzadeh
dc.date.accessioned2023-02-15T10:37:26Z
dc.date.available2023-02-15T10:37:26Z
dc.date.issued2022-07-23
dc.departmentDepartment of Computer Engineeringen_US
dc.description.abstractThe support vector data description (SVDD) approach serves as a de facto standard for one-class classification where the learning task entails inferring the smallest hyper-sphere to enclose target objects while linearly penalising the errors/slacks via an ℓ1-norm penalty term. In this study, we generalise this modelling formalism to a general ℓp-norm (p ≥ 1) penalty function on slacks. By virtue of an ℓp-norm function, in the primal space, the proposed approach enables formulating a non-linear cost for slacks. From a dual problem perspective, the proposed method introduces a dual norm into the objective function, thus, proving a controlling mechanism to tune into the intrinsic sparsity/uniformity of the problem for enhanced descriptive capability. A theoretical analysis based on Rademacher complexities characterises the generalisation performance of the proposed approach while the experimental results on several datasets confirm the merits of the proposed method compared to other alternatives.en_US
dc.description.provenanceSubmitted by Ezgi Uğurlu (ezgi.ugurlu@bilkent.edu.tr) on 2023-02-15T10:37:26Z No. of bitstreams: 1 ℓp-norm_support_vector_data_description.pdf: 1383202 bytes, checksum: 0fd4bd7e24f95ec8dfbed0d6c253072c (MD5)en
dc.description.provenanceMade available in DSpace on 2023-02-15T10:37:26Z (GMT). No. of bitstreams: 1 ℓp-norm_support_vector_data_description.pdf: 1383202 bytes, checksum: 0fd4bd7e24f95ec8dfbed0d6c253072c (MD5) Previous issue date: 2022-07-23en
dc.embargo.release2024-07-23
dc.identifier.doi10.1016/j.patcog.2022.108930en_US
dc.identifier.eissn1873-5142
dc.identifier.issn0031-3203
dc.identifier.urihttp://hdl.handle.net/11693/111326
dc.language.isoEnglishen_US
dc.publisherElsevier BVen_US
dc.relation.isversionofhttps://doi.org/10.1016/j.patcog.2022.108930en_US
dc.source.titlePattern Recognitionen_US
dc.subjectOne-class classificationen_US
dc.subjectKernel methodsen_US
dc.subjectSupport vector data descriptionen_US
dc.subjectℓp -norm penaltyen_US
dc.titleℓp-norm support vector data descriptionen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
ℓp-norm_support_vector_data_description.pdf
Size:
1.32 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.69 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Scholarly Publications - Computer Engineering