Matches in SemOpenAlex for { <https://semopenalex.org/work/W3134760956> ?p ?o ?g. }
- W3134760956 abstract "In the non-negative matrix factorization (NMF) problem, the input is an $mtimes n$ matrix $M$ with non-negative entries and the goal is to factorize it as $Mapprox AW$. The $mtimes k$ matrix $A$ and the $ktimes n$ matrix $W$ are both constrained to have non-negative entries. This is in contrast to singular value decomposition, where the matrices $A$ and $W$ can have negative entries but must satisfy the orthogonality constraint: the columns of $A$ are orthogonal and the rows of $W$ are also orthogonal. The orthogonal non-negative matrix factorization (ONMF) problem imposes both the non-negativity and the orthogonality constraints, and previous work showed that it leads to better performances than NMF on many clustering tasks. We give the first constant-factor approximation algorithm for ONMF when one or both of $A$ and $W$ are subject to the orthogonality constraint. We also show an interesting connection to the correlation clustering problem on bipartite graphs. Our experiments on synthetic and real-world data show that our algorithm achieves similar or smaller errors compared to previous ONMF algorithms while ensuring perfect orthogonality (many previous algorithms do not satisfy the hard orthogonality constraint)." @default.
- W3134760956 created "2021-03-15" @default.
- W3134760956 creator A5071016611 @default.
- W3134760956 creator A5076540826 @default.
- W3134760956 date "2021-03-02" @default.
- W3134760956 modified "2023-09-24" @default.
- W3134760956 title "Approximation Algorithms for Orthogonal Non-negative Matrix Factorization" @default.
- W3134760956 cites W1504886279 @default.
- W3134760956 cites W1511345785 @default.
- W3134760956 cites W1741870144 @default.
- W3134760956 cites W1763420192 @default.
- W3134760956 cites W1902027874 @default.
- W3134760956 cites W1970039276 @default.
- W3134760956 cites W1988153470 @default.
- W3134760956 cites W2004026774 @default.
- W3134760956 cites W2013029404 @default.
- W3134760956 cites W2030848960 @default.
- W3134760956 cites W2039030056 @default.
- W3134760956 cites W2043545458 @default.
- W3134760956 cites W2059745395 @default.
- W3134760956 cites W2063392856 @default.
- W3134760956 cites W2064980127 @default.
- W3134760956 cites W206759535 @default.
- W3134760956 cites W2089468765 @default.
- W3134760956 cites W2089917999 @default.
- W3134760956 cites W2091858563 @default.
- W3134760956 cites W2092612387 @default.
- W3134760956 cites W2093492509 @default.
- W3134760956 cites W2101139104 @default.
- W3134760956 cites W2110096996 @default.
- W3134760956 cites W2110672659 @default.
- W3134760956 cites W2121882762 @default.
- W3134760956 cites W2124172487 @default.
- W3134760956 cites W2132692097 @default.
- W3134760956 cites W2136171036 @default.
- W3134760956 cites W2136787567 @default.
- W3134760956 cites W2139258080 @default.
- W3134760956 cites W2145725490 @default.
- W3134760956 cites W2148541040 @default.
- W3134760956 cites W2165685007 @default.
- W3134760956 cites W2171125141 @default.
- W3134760956 cites W2189591224 @default.
- W3134760956 cites W2199495299 @default.
- W3134760956 cites W2333677034 @default.
- W3134760956 cites W2341301259 @default.
- W3134760956 cites W2507533531 @default.
- W3134760956 cites W2536611493 @default.
- W3134760956 cites W2565644440 @default.
- W3134760956 cites W2617081475 @default.
- W3134760956 cites W2802125049 @default.
- W3134760956 cites W2802183458 @default.
- W3134760956 cites W2803106003 @default.
- W3134760956 cites W2914959486 @default.
- W3134760956 cites W2955055966 @default.
- W3134760956 cites W2990947011 @default.
- W3134760956 cites W2992689909 @default.
- W3134760956 cites W3005704212 @default.
- W3134760956 cites W3120740533 @default.
- W3134760956 cites W3143596294 @default.
- W3134760956 hasPublicationYear "2021" @default.
- W3134760956 type Work @default.
- W3134760956 sameAs 3134760956 @default.
- W3134760956 citedByCount "0" @default.
- W3134760956 crossrefType "posted-content" @default.
- W3134760956 hasAuthorship W3134760956A5071016611 @default.
- W3134760956 hasAuthorship W3134760956A5076540826 @default.
- W3134760956 hasConcept C106487976 @default.
- W3134760956 hasConcept C11413529 @default.
- W3134760956 hasConcept C114614502 @default.
- W3134760956 hasConcept C118615104 @default.
- W3134760956 hasConcept C121332964 @default.
- W3134760956 hasConcept C132525143 @default.
- W3134760956 hasConcept C152671427 @default.
- W3134760956 hasConcept C158693339 @default.
- W3134760956 hasConcept C159985019 @default.
- W3134760956 hasConcept C17137986 @default.
- W3134760956 hasConcept C187834632 @default.
- W3134760956 hasConcept C188060507 @default.
- W3134760956 hasConcept C192562407 @default.
- W3134760956 hasConcept C197657726 @default.
- W3134760956 hasConcept C22789450 @default.
- W3134760956 hasConcept C2524010 @default.
- W3134760956 hasConcept C2776036281 @default.
- W3134760956 hasConcept C33923547 @default.
- W3134760956 hasConcept C42355184 @default.
- W3134760956 hasConcept C62520636 @default.
- W3134760956 hasConceptScore W3134760956C106487976 @default.
- W3134760956 hasConceptScore W3134760956C11413529 @default.
- W3134760956 hasConceptScore W3134760956C114614502 @default.
- W3134760956 hasConceptScore W3134760956C118615104 @default.
- W3134760956 hasConceptScore W3134760956C121332964 @default.
- W3134760956 hasConceptScore W3134760956C132525143 @default.
- W3134760956 hasConceptScore W3134760956C152671427 @default.
- W3134760956 hasConceptScore W3134760956C158693339 @default.
- W3134760956 hasConceptScore W3134760956C159985019 @default.
- W3134760956 hasConceptScore W3134760956C17137986 @default.
- W3134760956 hasConceptScore W3134760956C187834632 @default.
- W3134760956 hasConceptScore W3134760956C188060507 @default.
- W3134760956 hasConceptScore W3134760956C192562407 @default.
- W3134760956 hasConceptScore W3134760956C197657726 @default.