{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,6,13]],"date-time":"2024-06-13T01:35:30Z","timestamp":1718242530235},"reference-count":21,"publisher":"MIT Press - Journals","issue":"11","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Neural Computation"],"published-print":{"date-parts":[[2010,11]]},"abstract":"<jats:p> A variety of modifications have been employed to learning vector quantization (LVQ) algorithms using either crisp or soft windows for selection of data. Although these schemes have been shown in practice to improve performance, a theoretical study on the influence of windows has so far been limited. Here we rigorously analyze the influence of windows in a controlled environment of gaussian mixtures in high dimensions. Concepts from statistical physics and the theory of online learning allow an exact description of the training dynamics, yielding typical learning curves, convergence properties, and achievable generalization abilities. We compare the performance and demonstrate the advantages of various algorithms, including LVQ 2.1, generalized LVQ (GLVQ), Learning from Mistakes (LFM) and Robust Soft LVQ (RSLVQ). We find that the selection of the window parameter highly influences the learning curves but not, surprisingly, the asymptotic performances of LVQ 2.1 and RSLVQ. Although the prototypes of LVQ 2.1 exhibit divergent behavior, the resulting decision boundary coincides with the optimal decision boundary, thus yielding optimal generalization ability. <\/jats:p>","DOI":"10.1162\/neco_a_00030","type":"journal-article","created":{"date-parts":[[2010,8,30]],"date-time":"2010-08-30T15:43:54Z","timestamp":1283183034000},"page":"2924-2961","source":"Crossref","is-referenced-by-count":11,"title":["Window-Based Example Selection in Learning Vector Quantization"],"prefix":"10.1162","volume":"22","author":[{"given":"A. W.","family":"Witoelar","sequence":"first","affiliation":[{"name":"Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen, Groningen 9747 AG, Netherlands"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"A.","family":"Ghosh","sequence":"additional","affiliation":[{"name":"School of Computer Science and Statistics, Trinity College, Dublin 2, Ireland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"J. J. G.","family":"de Vries","sequence":"additional","affiliation":[{"name":"Philips Research Europe, Eindhoven 5656 AE, Netherlands"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"B.","family":"Hammer","sequence":"additional","affiliation":[{"name":"Center of Excellence Cognitive Interaction Technology, Bielefeld University, 33615 Bielefeld, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"M.","family":"Biehl","sequence":"additional","affiliation":[{"name":"Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen, Groningen 9747 AG, Netherlands"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.70.3167"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1162\/089976600300015187"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1209\/0295-5075\/25\/5\/014"},{"key":"B4","first-page":"1095","volume-title":"The handbook of brain theory and neural networks","author":"Biehl M.","year":"2003"},{"key":"B6","first-page":"323","volume":"8","author":"Biehl M.","year":"2007","journal-title":"J. Mach. Learning Res."},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1209\/0295-5075\/24\/5\/017"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1088\/0305-4470\/26\/11\/014"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9781139164542"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2006.05.010"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(02)00079-5"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.1990.137622"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-56927-2"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1995.7.1.144"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.80.5445"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511569920"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.79.2578"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.52.4225"},{"key":"B20","first-page":"423","volume-title":"Advances in neural information processing systems","volume":"7","author":"Sato A.","year":"1995"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1162\/089976603321891819"},{"key":"B22","volume-title":"International Joint Conference on Neural Networks","author":"Seo S.","year":"2006"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2007.11.022"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mitpressjournals.org\/doi\/pdf\/10.1162\/NECO_a_00030","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,12]],"date-time":"2021-03-12T21:38:22Z","timestamp":1615585102000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/22\/11\/2924-2961\/7594"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2010,11]]},"references-count":21,"journal-issue":{"issue":"11","published-print":{"date-parts":[[2010,11]]}},"alternative-id":["10.1162\/NECO_a_00030"],"URL":"https:\/\/doi.org\/10.1162\/neco_a_00030","relation":{},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2010,11]]}}}