Matches in SemOpenAlex for { <https://semopenalex.org/work/W3058957010> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W3058957010 abstract "Multi-category support vector machine (MC-SVM) is one of the most popular machine learning algorithms. There are lots of variants of MC-SVM, although different optimization algorithms were developed for different learning machines. In this study, we developed a new optimization algorithm that can be applied to many of MC-SVM variants. The algorithm is based on the Frank-Wolfe framework that requires two subproblems, direction finding and line search, in each iteration. The contribution of this study is the discovery that both subproblems have a closed form solution if the Frank-Wolfe framework is applied to the dual problem. Additionally, the closed form solutions on both for the direction finding and for the line search exist even for the Moreau envelopes of the loss functions. We use several large datasets to demonstrate that the proposed optimization algorithm converges rapidly and thereby improves the pattern recognition performance." @default.
- W3058957010 created "2020-08-24" @default.
- W3058957010 creator A5003163282 @default.
- W3058957010 creator A5011841010 @default.
- W3058957010 creator A5031832499 @default.
- W3058957010 creator A5061978732 @default.
- W3058957010 date "2020-08-20" @default.
- W3058957010 modified "2023-09-27" @default.
- W3058957010 title "Frank-Wolfe algorithm for learning SVM-type multi-category classifiers" @default.
- W3058957010 cites W1939652453 @default.
- W3058957010 cites W1987063155 @default.
- W3058957010 cites W2031248101 @default.
- W3058957010 cites W2046033161 @default.
- W3058957010 cites W2070771761 @default.
- W3058957010 cites W2105842272 @default.
- W3058957010 cites W2136885855 @default.
- W3058957010 cites W2157791002 @default.
- W3058957010 cites W2168356304 @default.
- W3058957010 cites W2343187456 @default.
- W3058957010 cites W2798766386 @default.
- W3058957010 cites W2804776105 @default.
- W3058957010 cites W2963899139 @default.
- W3058957010 cites W2990448268 @default.
- W3058957010 hasPublicationYear "2020" @default.
- W3058957010 type Work @default.
- W3058957010 sameAs 3058957010 @default.
- W3058957010 citedByCount "0" @default.
- W3058957010 crossrefType "posted-content" @default.
- W3058957010 hasAuthorship W3058957010A5003163282 @default.
- W3058957010 hasAuthorship W3058957010A5011841010 @default.
- W3058957010 hasAuthorship W3058957010A5031832499 @default.
- W3058957010 hasAuthorship W3058957010A5061978732 @default.
- W3058957010 hasConcept C11413529 @default.
- W3058957010 hasConcept C119857082 @default.
- W3058957010 hasConcept C12267149 @default.
- W3058957010 hasConcept C124952713 @default.
- W3058957010 hasConcept C124975894 @default.
- W3058957010 hasConcept C142362112 @default.
- W3058957010 hasConcept C154945302 @default.
- W3058957010 hasConcept C198352243 @default.
- W3058957010 hasConcept C2524010 @default.
- W3058957010 hasConcept C2780980858 @default.
- W3058957010 hasConcept C33923547 @default.
- W3058957010 hasConcept C41008148 @default.
- W3058957010 hasConceptScore W3058957010C11413529 @default.
- W3058957010 hasConceptScore W3058957010C119857082 @default.
- W3058957010 hasConceptScore W3058957010C12267149 @default.
- W3058957010 hasConceptScore W3058957010C124952713 @default.
- W3058957010 hasConceptScore W3058957010C124975894 @default.
- W3058957010 hasConceptScore W3058957010C142362112 @default.
- W3058957010 hasConceptScore W3058957010C154945302 @default.
- W3058957010 hasConceptScore W3058957010C198352243 @default.
- W3058957010 hasConceptScore W3058957010C2524010 @default.
- W3058957010 hasConceptScore W3058957010C2780980858 @default.
- W3058957010 hasConceptScore W3058957010C33923547 @default.
- W3058957010 hasConceptScore W3058957010C41008148 @default.
- W3058957010 hasLocation W30589570101 @default.
- W3058957010 hasOpenAccess W3058957010 @default.
- W3058957010 hasPrimaryLocation W30589570101 @default.
- W3058957010 hasRelatedWork W1511131371 @default.
- W3058957010 hasRelatedWork W1988872058 @default.
- W3058957010 hasRelatedWork W1997405810 @default.
- W3058957010 hasRelatedWork W2029384563 @default.
- W3058957010 hasRelatedWork W2074013573 @default.
- W3058957010 hasRelatedWork W2077808175 @default.
- W3058957010 hasRelatedWork W2113159480 @default.
- W3058957010 hasRelatedWork W2368947495 @default.
- W3058957010 hasRelatedWork W2373716234 @default.
- W3058957010 hasRelatedWork W2465334159 @default.
- W3058957010 hasRelatedWork W2505554796 @default.
- W3058957010 hasRelatedWork W2570835487 @default.
- W3058957010 hasRelatedWork W2657468497 @default.
- W3058957010 hasRelatedWork W2944689333 @default.
- W3058957010 hasRelatedWork W2951153209 @default.
- W3058957010 hasRelatedWork W2951858610 @default.
- W3058957010 hasRelatedWork W3016454111 @default.
- W3058957010 hasRelatedWork W3035360656 @default.
- W3058957010 hasRelatedWork W3184230123 @default.
- W3058957010 hasRelatedWork W3197426870 @default.
- W3058957010 isParatext "false" @default.
- W3058957010 isRetracted "false" @default.
- W3058957010 magId "3058957010" @default.
- W3058957010 workType "article" @default.