Adaptive relevance matrices in learning vector quantization
We propose a new matrix learning scheme to extend relevance learning vector quantization (RLVQ), an efficient prototype-based classification algorithm, toward a general adaptive metric. By introducing a full matrix of relevance factors in the distance measure, correlations between different features...
Saved in:
| Published in: | Neural computation Vol. 21; no. 12; p. 3532 |
|---|---|
| Main Authors: | , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
United States
01.12.2009
|
| Subjects: | |
| ISSN: | 0899-7667 |
| Online Access: | Get more information |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | We propose a new matrix learning scheme to extend relevance learning vector quantization (RLVQ), an efficient prototype-based classification algorithm, toward a general adaptive metric. By introducing a full matrix of relevance factors in the distance measure, correlations between different features and their importance for the classification scheme can be taken into account and automated, and general metric adaptation takes place during training. In comparison to the weighted Euclidean metric used in RLVQ and its variations, a full matrix is more powerful to represent the internal structure of the data appropriately. Large margin generalization bounds can be transferred to this case, leading to bounds that are independent of the input dimensionality. This also holds for local metrics attached to each prototype, which corresponds to piecewise quadratic decision boundaries. The algorithm is tested in comparison to alternative learning vector quantization schemes using an artificial data set, a benchmark multiclass problem from the UCI repository, and a problem from bioinformatics, the recognition of splice sites for C. elegans. |
|---|---|
| AbstractList | We propose a new matrix learning scheme to extend relevance learning vector quantization (RLVQ), an efficient prototype-based classification algorithm, toward a general adaptive metric. By introducing a full matrix of relevance factors in the distance measure, correlations between different features and their importance for the classification scheme can be taken into account and automated, and general metric adaptation takes place during training. In comparison to the weighted Euclidean metric used in RLVQ and its variations, a full matrix is more powerful to represent the internal structure of the data appropriately. Large margin generalization bounds can be transferred to this case, leading to bounds that are independent of the input dimensionality. This also holds for local metrics attached to each prototype, which corresponds to piecewise quadratic decision boundaries. The algorithm is tested in comparison to alternative learning vector quantization schemes using an artificial data set, a benchmark multiclass problem from the UCI repository, and a problem from bioinformatics, the recognition of splice sites for C. elegans.We propose a new matrix learning scheme to extend relevance learning vector quantization (RLVQ), an efficient prototype-based classification algorithm, toward a general adaptive metric. By introducing a full matrix of relevance factors in the distance measure, correlations between different features and their importance for the classification scheme can be taken into account and automated, and general metric adaptation takes place during training. In comparison to the weighted Euclidean metric used in RLVQ and its variations, a full matrix is more powerful to represent the internal structure of the data appropriately. Large margin generalization bounds can be transferred to this case, leading to bounds that are independent of the input dimensionality. This also holds for local metrics attached to each prototype, which corresponds to piecewise quadratic decision boundaries. The algorithm is tested in comparison to alternative learning vector quantization schemes using an artificial data set, a benchmark multiclass problem from the UCI repository, and a problem from bioinformatics, the recognition of splice sites for C. elegans. We propose a new matrix learning scheme to extend relevance learning vector quantization (RLVQ), an efficient prototype-based classification algorithm, toward a general adaptive metric. By introducing a full matrix of relevance factors in the distance measure, correlations between different features and their importance for the classification scheme can be taken into account and automated, and general metric adaptation takes place during training. In comparison to the weighted Euclidean metric used in RLVQ and its variations, a full matrix is more powerful to represent the internal structure of the data appropriately. Large margin generalization bounds can be transferred to this case, leading to bounds that are independent of the input dimensionality. This also holds for local metrics attached to each prototype, which corresponds to piecewise quadratic decision boundaries. The algorithm is tested in comparison to alternative learning vector quantization schemes using an artificial data set, a benchmark multiclass problem from the UCI repository, and a problem from bioinformatics, the recognition of splice sites for C. elegans. |
| Author | Schneider, Petra Biehl, Michael Hammer, Barbara |
| Author_xml | – sequence: 1 givenname: Petra surname: Schneider fullname: Schneider, Petra email: p.schneider@rug.nl organization: Institute of Mathematics and Computing Science, University of Groningen, 9700 AK Groningen, The Netherlands. p.schneider@rug.nl – sequence: 2 givenname: Michael surname: Biehl fullname: Biehl, Michael – sequence: 3 givenname: Barbara surname: Hammer fullname: Hammer, Barbara |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/19764875$$D View this record in MEDLINE/PubMed |
| BookMark | eNo1j0tLxDAURrMYcR76CwTpzlXHmybNA1fD4AsG3Oi6JOmtRNp0JkkH9Nc74Lg6fHD44CzJLIwBCbmhsKZUVPcB3biuAPRplqBKDWpGFqC0LqUQck6WKX0BgKBQX5I51VJwJesFedi0Zp_9EYuIPR5NcFgMJkfvMBU-FD2aGHz4LI7o8hiLw2RC9j8m-zFckYvO9Amvz1yRj6fH9-1LuXt7ft1udqXjWuXSWqGs5lJp5JwDKEFbxowUbVUrpasKdeeEZdqorhZoLQXDmIZOUlYjxWpF7v5-93E8TJhyM_jksO9NwHFKjWScMiUpnMzbsznZAdtmH_1g4nfz31v9AnO6V-s |
| CitedBy_id | crossref_primary_10_1016_j_neucom_2020_03_025 crossref_primary_10_1007_s00521_013_1535_3 crossref_primary_10_1016_j_ic_2017_08_004 crossref_primary_10_1016_j_neucom_2014_03_003 crossref_primary_10_1016_j_patcog_2024_110497 crossref_primary_10_1016_j_compbiomed_2025_110180 crossref_primary_10_1007_s11063_013_9278_9 crossref_primary_10_1016_j_neucom_2014_11_082 crossref_primary_10_1109_TCYB_2022_3178412 crossref_primary_10_1016_j_cie_2024_110272 crossref_primary_10_3389_fncom_2016_00117 crossref_primary_10_1136_bmjopen_2021_055068 crossref_primary_10_1007_s00259_025_07114_4 crossref_primary_10_1016_j_neucom_2011_09_031 crossref_primary_10_1007_s42979_023_01782_5 crossref_primary_10_1017_S1743921316012928 crossref_primary_10_1016_j_isprsjprs_2010_11_001 crossref_primary_10_1016_j_patcog_2016_10_011 crossref_primary_10_1515_jaiscr_2017_0005 crossref_primary_10_1109_TETCI_2024_3515009 crossref_primary_10_1016_j_neucom_2017_01_105 crossref_primary_10_3390_s22249689 crossref_primary_10_1007_s13369_024_08742_y crossref_primary_10_1515_auto_2017_0123 crossref_primary_10_1016_j_neucom_2017_11_072 crossref_primary_10_1016_j_neucom_2018_11_095 crossref_primary_10_1007_s41060_018_0163_5 crossref_primary_10_1016_j_neucom_2017_11_074 crossref_primary_10_1016_j_neucom_2023_126632 crossref_primary_10_1371_journal_pone_0059401 crossref_primary_10_1007_s00521_021_06018_2 crossref_primary_10_1016_j_artmed_2012_07_001 crossref_primary_10_1080_19393555_2024_2378747 crossref_primary_10_1016_j_geoderma_2019_01_018 crossref_primary_10_1210_clinem_dgae604 crossref_primary_10_1016_j_neucom_2015_12_108 crossref_primary_10_1016_j_neucom_2015_12_109 crossref_primary_10_1016_j_neucom_2017_06_084 crossref_primary_10_1016_j_neunet_2021_04_024 crossref_primary_10_3390_s21134405 crossref_primary_10_1007_s00521_018_03966_0 crossref_primary_10_1016_j_neucom_2013_07_049 crossref_primary_10_1007_s12652_020_02218_1 crossref_primary_10_1016_j_renene_2016_05_092 crossref_primary_10_1007_s00500_014_1496_1 crossref_primary_10_3390_app9194036 crossref_primary_10_1109_TIP_2019_2948472 crossref_primary_10_1016_j_neucom_2011_11_029 crossref_primary_10_1007_s00521_020_05517_y crossref_primary_10_1007_s13218_015_0372_1 crossref_primary_10_1007_s11063_014_9394_1 crossref_primary_10_1007_s12541_017_0113_5 crossref_primary_10_3390_e25030540 crossref_primary_10_1080_08839514_2023_2198846 crossref_primary_10_1016_j_patcog_2010_10_024 crossref_primary_10_1038_s41598_025_94692_w crossref_primary_10_1109_TPAMI_2024_3466315 crossref_primary_10_1136_annrheumdis_2014_206921 crossref_primary_10_1007_s42979_024_03067_x crossref_primary_10_1016_j_neucom_2024_127367 crossref_primary_10_1109_TNSRE_2019_2907200 crossref_primary_10_1007_s00138_016_0781_7 crossref_primary_10_1016_j_neucom_2019_12_131 crossref_primary_10_1016_j_artmed_2013_11_003 crossref_primary_10_1007_s40846_017_0353_y crossref_primary_10_1162_NECO_a_00358 crossref_primary_10_3390_e23101357 crossref_primary_10_1016_j_nicl_2020_102199 crossref_primary_10_1016_j_neucom_2014_12_002 crossref_primary_10_1007_s11831_025_10267_y crossref_primary_10_1016_j_imavis_2016_10_006 crossref_primary_10_1007_s00521_019_04299_2 crossref_primary_10_1016_j_neucom_2013_11_049 crossref_primary_10_1016_j_pmcj_2023_101752 crossref_primary_10_1109_TCBB_2014_2377750 crossref_primary_10_1210_jc_2011_1565 crossref_primary_10_1016_j_neucom_2013_11_048 crossref_primary_10_1007_s00521_019_04080_5 crossref_primary_10_1016_j_neucom_2024_128100 crossref_primary_10_1162_NECO_a_00872 crossref_primary_10_1016_j_neucom_2013_05_054 crossref_primary_10_1016_j_neucom_2009_11_017 crossref_primary_10_1016_j_neucom_2010_12_015 crossref_primary_10_1016_j_artmed_2024_102786 crossref_primary_10_1162_NECO_a_00250 crossref_primary_10_1186_s12888_025_06536_6 crossref_primary_10_1016_j_neucom_2021_05_105 crossref_primary_10_1016_j_neucom_2014_12_096 crossref_primary_10_1109_TNSRE_2023_3295453 crossref_primary_10_1016_j_neucom_2022_06_035 crossref_primary_10_1016_j_cmpb_2020_105708 crossref_primary_10_1109_ACCESS_2021_3087231 crossref_primary_10_1007_s11265_020_01520_7 crossref_primary_10_1016_j_neunet_2017_05_006 crossref_primary_10_1080_03081079_2012_723209 crossref_primary_10_1016_j_neunet_2011_10_001 crossref_primary_10_1109_TNNLS_2013_2251470 crossref_primary_10_1016_j_procs_2025_07_080 crossref_primary_10_1016_j_neucom_2023_02_040 crossref_primary_10_1016_j_jsbmb_2023_106445 crossref_primary_10_1016_j_eswa_2015_04_034 crossref_primary_10_1007_s10851_012_0356_9 crossref_primary_10_1016_j_neucom_2010_10_016 crossref_primary_10_1016_j_neucom_2011_10_018 crossref_primary_10_1002_wcs_1378 crossref_primary_10_1016_j_neunet_2022_07_004 crossref_primary_10_1016_j_cmpb_2022_107042 crossref_primary_10_1109_TNNLS_2020_2978514 crossref_primary_10_1016_j_neunet_2011_05_013 crossref_primary_10_1002_2050_7038_13046 crossref_primary_10_1016_j_neucom_2014_10_092 crossref_primary_10_1007_s00180_016_0678_y crossref_primary_10_1109_ACCESS_2018_2890544 crossref_primary_10_1007_s13218_012_0188_1 crossref_primary_10_1109_TSMC_2015_2468192 crossref_primary_10_1109_ACCESS_2019_2947581 crossref_primary_10_1111_apt_15710 crossref_primary_10_1109_TSMC_2017_2761360 crossref_primary_10_1016_j_neucom_2014_10_008 crossref_primary_10_1038_s41467_022_28795_7 crossref_primary_10_1016_j_ascom_2022_100555 crossref_primary_10_1007_s11227_018_2684_z crossref_primary_10_1016_j_neucom_2025_129405 crossref_primary_10_1016_j_neucom_2016_06_038 crossref_primary_10_1016_j_neunet_2012_04_010 crossref_primary_10_1016_j_neucom_2018_12_076 crossref_primary_10_1371_journal_pone_0117157 crossref_primary_10_1007_s00787_021_01893_5 crossref_primary_10_1109_TNN_2010_2042729 crossref_primary_10_1016_j_neucom_2021_04_129 crossref_primary_10_1016_j_tcs_2018_04_045 |
| ContentType | Journal Article |
| DBID | CGR CUY CVF ECM EIF NPM 7X8 |
| DOI | 10.1162/neco.2009.11-08-908 |
| DatabaseName | Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic |
| DatabaseTitle | MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic MEDLINE |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | no_fulltext_linktorsrc |
| Discipline | Computer Science |
| ExternalDocumentID | 19764875 |
| Genre | Journal Article |
| GroupedDBID | --- -~X .4S .DC 0R~ 123 36B 4.4 41~ 53G 6IK AAFWJ AAJGR AALMD ABAZT ABDBF ABDNZ ABEFU ABIVO ABJNI ABVLG ACGFO ACUHS ACYGS ADIYS ADMLS AEGXH AEILP AENEX AIAGR ALMA_UNASSIGNED_HOLDINGS AMVHM ARCSS AVWKF AZFZN BEFXN BFFAM BGNUA BKEBE BPEOZ CAG CGR COF CS3 CUY CVF DU5 EAP EAS EBC EBD EBS ECM ECS EDO EIF EJD EMB EMK EMOBN EPL EPS EST ESX F5P FEDTE FNEHJ HVGLF HZ~ H~9 I-F IPLJI JAVBF MCG MINIK MKJ NPM O9- OCL P2P PK0 PQQKQ RMI SV3 TUS WG8 WH7 XJE ZWS 7X8 |
| ID | FETCH-LOGICAL-c498t-bb68b94789e44400861d33a76d2588922e9fc6b39a8f56ebb10a3390f7135e1e2 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 226 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000272200100011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0899-7667 |
| IngestDate | Thu Sep 04 20:21:17 EDT 2025 Mon Jul 21 06:02:10 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 12 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c498t-bb68b94789e44400861d33a76d2588922e9fc6b39a8f56ebb10a3390f7135e1e2 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| PMID | 19764875 |
| PQID | 734138710 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_734138710 pubmed_primary_19764875 |
| PublicationCentury | 2000 |
| PublicationDate | 2009-12-01 |
| PublicationDateYYYYMMDD | 2009-12-01 |
| PublicationDate_xml | – month: 12 year: 2009 text: 2009-12-01 day: 01 |
| PublicationDecade | 2000 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States |
| PublicationTitle | Neural computation |
| PublicationTitleAlternate | Neural Comput |
| PublicationYear | 2009 |
| SSID | ssj0006105 |
| Score | 2.4611006 |
| Snippet | We propose a new matrix learning scheme to extend relevance learning vector quantization (RLVQ), an efficient prototype-based classification algorithm, toward... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| StartPage | 3532 |
| SubjectTerms | Algorithms Artificial Intelligence Automatic Data Processing Brain - physiology Humans Information Theory Learning - physiology Probability Learning |
| Title | Adaptive relevance matrices in learning vector quantization |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/19764875 https://www.proquest.com/docview/734138710 |
| Volume | 21 |
| WOSCitedRecordID | wos000272200100011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV05T8MwFLaAMrBwH-WSB1ar8VEfYkAVomKh6gBSt8gn6kDakra_HztxYUIMLNkSRS9-L9_3_Px9ANwRTbA1ItJU5w1iLuacsjggmaTwIiNi2LVmE2I0kpOJGufZnDqPVW5qYlOo3cymHnlPpHIb0X3xMF-gZBqVNlezg8Y26NCIZFJeismPWDjPE4yRUiDBuciiQ5iTXhW5XStW2XYIVSF_h5jNr2Z48M-XPAT7GWPCQbsojsCWr47Bwca_AeZ0PgH3A6fnqdzB5JzSzALAj0ay39dwWsHsKPEO101rHy5W8TPkc5un4G349Pr4jLKZArJMySUyhkujmJDKM8YSk8GOUi24I30pFSFeBcsNVVqGPvfG4EJTqoqQPPw89uQM7FSzyl-kY94KW-4LZhVlnmvpWOA8PoIRFyJg6AK4iU4ZF2vagdCVn63q8js-XXDeRrict6IaJY64KJGny79vvgJ7JLs4FPgadEJMVH8Ddu16Oa0_b5tFEK-j8csXtZm6Zg |
| linkProvider | ProQuest |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Adaptive+relevance+matrices+in+learning+vector+quantization&rft.jtitle=Neural+computation&rft.au=Schneider%2C+Petra&rft.au=Biehl%2C+Michael&rft.au=Hammer%2C+Barbara&rft.date=2009-12-01&rft.issn=0899-7667&rft.volume=21&rft.issue=12&rft.spage=3532&rft_id=info:doi/10.1162%2Fneco.2009.11-08-908&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0899-7667&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0899-7667&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0899-7667&client=summon |