Universal priors for sparse modeling
Resumen:
Sparse data models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and their use has led to state-of-the-art results in many signal and image processing tasks. It is now well understood that the choice of the sparsity regularization term is critical in the success of such models. In this work, we use tools from information theory to propose a sparsity regularization term which has several theoretical and practical advantages over the more standard ¿ 0 or ¿ 1 ones, and which leads to improved coding performance and accuracy in reconstruction tasks. We also briefly report on further improvements obtained by imposing low mutual coherence and Gram matrix norm on the learned dictionaries
| 2009 | |
| Inglés | |
| Universidad de la República | |
| COLIBRI | |
| https://hdl.handle.net/20.500.12008/38679 | |
| Acceso abierto | |
| Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) |
| _version_ | 1807522934226419712 |
|---|---|
| author | Ramírez Paulino, Ignacio |
| author2 | Lecumberry, Federico Sapiro, Guillermo |
| author2_role | author author |
| author_facet | Ramírez Paulino, Ignacio Lecumberry, Federico Sapiro, Guillermo |
| author_role | author |
| bitstream.checksum.fl_str_mv | 7f2e2c17ef6585de66da58d1bfa8b5e1 9833653f73f7853880c94a6fead477b1 4afdbb8c545fd630ea7db775da747b2f 9da0b6dfac957114c6a7714714b86306 acf78ce1e530254408390888497f4d66 |
| bitstream.checksumAlgorithm.fl_str_mv | MD5 MD5 MD5 MD5 MD5 |
| bitstream.url.fl_str_mv | http://localhost:8080/xmlui/bitstream/20.500.12008/38679/5/license.txt http://localhost:8080/xmlui/bitstream/20.500.12008/38679/2/license_text http://localhost:8080/xmlui/bitstream/20.500.12008/38679/3/license_url http://localhost:8080/xmlui/bitstream/20.500.12008/38679/4/license_rdf http://localhost:8080/xmlui/bitstream/20.500.12008/38679/1/RLS09.pdf |
| collection | COLIBRI |
| dc.creator.none.fl_str_mv | Ramírez Paulino, Ignacio Lecumberry, Federico Sapiro, Guillermo |
| dc.date.accessioned.none.fl_str_mv | 2023-08-01T20:33:18Z |
| dc.date.available.none.fl_str_mv | 2023-08-01T20:33:18Z |
| dc.date.issued.es.fl_str_mv | 2009 |
| dc.date.submitted.es.fl_str_mv | 20230801 |
| dc.description.abstract.none.fl_txt_mv | Sparse data models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and their use has led to state-of-the-art results in many signal and image processing tasks. It is now well understood that the choice of the sparsity regularization term is critical in the success of such models. In this work, we use tools from information theory to propose a sparsity regularization term which has several theoretical and practical advantages over the more standard ¿ 0 or ¿ 1 ones, and which leads to improved coding performance and accuracy in reconstruction tasks. We also briefly report on further improvements obtained by imposing low mutual coherence and Gram matrix norm on the learned dictionaries |
| dc.identifier.citation.es.fl_str_mv | Ramírez Paulino, I, Lecumberry, F, Sapiro, G. “Universal priors for sparse modeling”. Proceedings of the 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), Aruba, Dutch Antilles, 2009.. DOI: 10.1109/CAMSAP.2009.5413302 |
| dc.identifier.doi.es.fl_str_mv | DOI: 10.1109/CAMSAP.2009.5413302 |
| dc.identifier.uri.none.fl_str_mv | https://hdl.handle.net/20.500.12008/38679 |
| dc.language.iso.none.fl_str_mv | en eng |
| dc.publisher.es.fl_str_mv | IEEE |
| dc.relation.ispartof.es.fl_str_mv | 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), Aruba, Dutch Antilles, 2009. |
| dc.rights.license.none.fl_str_mv | Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) |
| dc.rights.none.fl_str_mv | info:eu-repo/semantics/openAccess |
| dc.source.none.fl_str_mv | reponame:COLIBRI instname:Universidad de la República instacron:Universidad de la República |
| dc.title.none.fl_str_mv | Universal priors for sparse modeling |
| dc.type.es.fl_str_mv | Ponencia |
| dc.type.none.fl_str_mv | info:eu-repo/semantics/conferenceObject |
| dc.type.version.none.fl_str_mv | info:eu-repo/semantics/publishedVersion |
| description | Sparse data models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and their use has led to state-of-the-art results in many signal and image processing tasks. It is now well understood that the choice of the sparsity regularization term is critical in the success of such models. In this work, we use tools from information theory to propose a sparsity regularization term which has several theoretical and practical advantages over the more standard ¿ 0 or ¿ 1 ones, and which leads to improved coding performance and accuracy in reconstruction tasks. We also briefly report on further improvements obtained by imposing low mutual coherence and Gram matrix norm on the learned dictionaries |
| eu_rights_str_mv | openAccess |
| format | conferenceObject |
| id | COLIBRI_9ac6e6fb5431e353f23befe01d331ce1 |
| identifier_str_mv | Ramírez Paulino, I, Lecumberry, F, Sapiro, G. “Universal priors for sparse modeling”. Proceedings of the 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), Aruba, Dutch Antilles, 2009.. DOI: 10.1109/CAMSAP.2009.5413302 DOI: 10.1109/CAMSAP.2009.5413302 |
| instacron_str | Universidad de la República |
| institution | Universidad de la República |
| instname_str | Universidad de la República |
| language | eng |
| language_invalid_str_mv | en |
| network_acronym_str | COLIBRI |
| network_name_str | COLIBRI |
| oai_identifier_str | oai:colibri.udelar.edu.uy:20.500.12008/38679 |
| publishDate | 2009 |
| reponame_str | COLIBRI |
| repository.mail.fl_str_mv | mabel.seroubian@seciu.edu.uy |
| repository.name.fl_str_mv | COLIBRI - Universidad de la República |
| repository_id_str | 4771 |
| rights_invalid_str_mv | Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) |
| spelling | 2023-08-01T20:33:18Z2023-08-01T20:33:18Z200920230801Ramírez Paulino, I, Lecumberry, F, Sapiro, G. “Universal priors for sparse modeling”. Proceedings of the 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), Aruba, Dutch Antilles, 2009.. DOI: 10.1109/CAMSAP.2009.5413302https://hdl.handle.net/20.500.12008/38679DOI: 10.1109/CAMSAP.2009.5413302Sparse data models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and their use has led to state-of-the-art results in many signal and image processing tasks. It is now well understood that the choice of the sparsity regularization term is critical in the success of such models. In this work, we use tools from information theory to propose a sparsity regularization term which has several theoretical and practical advantages over the more standard ¿ 0 or ¿ 1 ones, and which leads to improved coding performance and accuracy in reconstruction tasks. We also briefly report on further improvements obtained by imposing low mutual coherence and Gram matrix norm on the learned dictionariesMade available in DSpace on 2023-08-01T20:33:18Z (GMT). No. of bitstreams: 5 RLS09.pdf: 621827 bytes, checksum: acf78ce1e530254408390888497f4d66 (MD5) license_text: 21936 bytes, checksum: 9833653f73f7853880c94a6fead477b1 (MD5) license_url: 49 bytes, checksum: 4afdbb8c545fd630ea7db775da747b2f (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) license.txt: 4194 bytes, checksum: 7f2e2c17ef6585de66da58d1bfa8b5e1 (MD5) Previous issue date: 2009enengIEEE3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), Aruba, Dutch Antilles, 2009.Las obras depositadas en el Repositorio se rigen por la Ordenanza de los Derechos de la Propiedad Intelectual de la Universidad De La República. (Res. Nº 91 de C.D.C. de 8/III/1994 – D.O. 7/IV/1994) y por la Ordenanza del Repositorio Abierto de la Universidad de la República (Res. Nº 16 de C.D.C. de 07/10/2014)info:eu-repo/semantics/openAccessLicencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0)Universal priors for sparse modelingPonenciainfo:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/publishedVersionreponame:COLIBRIinstname:Universidad de la Repúblicainstacron:Universidad de la RepúblicaRamírez Paulino, IgnacioLecumberry, FedericoSapiro, GuillermoLICENSElicense.txttext/plain4194http://localhost:8080/xmlui/bitstream/20.500.12008/38679/5/license.txt7f2e2c17ef6585de66da58d1bfa8b5e1MD55CC-LICENSElicense_textapplication/octet-stream21936http://localhost:8080/xmlui/bitstream/20.500.12008/38679/2/license_text9833653f73f7853880c94a6fead477b1MD52license_urlapplication/octet-stream49http://localhost:8080/xmlui/bitstream/20.500.12008/38679/3/license_url4afdbb8c545fd630ea7db775da747b2fMD53license_rdfapplication/octet-stream23148http://localhost:8080/xmlui/bitstream/20.500.12008/38679/4/license_rdf9da0b6dfac957114c6a7714714b86306MD54ORIGINALRLS09.pdfapplication/pdf621827http://localhost:8080/xmlui/bitstream/20.500.12008/38679/1/RLS09.pdfacf78ce1e530254408390888497f4d66MD5120.500.12008/386792023-08-01 17:33:18.677oai:colibri.udelar.edu.uy:20.500.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://udelar.edu.uy/https://www.colibri.udelar.edu.uy/oai/requestmabel.seroubian@seciu.edu.uyUruguayopendoar:47712024-07-25T14:33:25.860131COLIBRI - Universidad de la Repúblicafalse |
| spellingShingle | Universal priors for sparse modeling Ramírez Paulino, Ignacio |
| status_str | publishedVersion |
| title | Universal priors for sparse modeling |
| title_full | Universal priors for sparse modeling |
| title_fullStr | Universal priors for sparse modeling |
| title_full_unstemmed | Universal priors for sparse modeling |
| title_short | Universal priors for sparse modeling |
| title_sort | Universal priors for sparse modeling |
| url | https://hdl.handle.net/20.500.12008/38679 |
