FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950646475> ?p ?o ?g. }

• W2950646475 abstract "An instance-weighted variant of the support vector machine (SVM) has attracted considerable attention recently since they are useful in various machine learning tasks such as non-stationary data analysis, heteroscedastic data modeling, transfer learning, learning to rank, and transduction. An important challenge in these scenarios is to overcome the computational bottleneck---instance weights often change dynamically or adaptively, and thus the weighted SVM solutions must be repeatedly computed. In this paper, we develop an algorithm that can efficiently and exactly update the weighted SVM solutions for arbitrary change of instance weights. Technically, this contribution can be regarded as an extension of the conventional solution-path algorithm for a single regularization parameter to multiple instance-weight parameters. However, this extension gives rise to a significant problem that breakpoints (at which the solution path turns) have to be identified in high-dimensional space. To facilitate this, we introduce a parametric representation of instance weights. We also provide a geometric interpretation in weight space using a notion of critical region: a polyhedron in which the current affine solution remains to be optimal. Then we find breakpoints at intersections of the solution path and boundaries of polyhedrons. Through extensive experiments on various practical applications, we demonstrate the usefulness of the proposed algorithm." @default.
• W2950646475 created "2019-06-27" @default.
• W2950646475 creator A5003668143 @default.
• W2950646475 creator A5042930568 @default.
• W2950646475 creator A5072744508 @default.
• W2950646475 creator A5081482638 @default.
• W2950646475 date "2010-09-24" @default.
• W2950646475 modified "2023-09-24" @default.
• W2950646475 title "Multi-parametric Solution-path Algorithm for Instance-weighted Support Vector Machines" @default.
• W2950646475 cites W107875978 @default.
• W2950646475 cites W1512098439 @default.
• W2950646475 cites W1556684468 @default.
• W2950646475 cites W1572899790 @default.
• W2950646475 cites W157995801 @default.
• W2950646475 cites W1607624180 @default.
• W2950646475 cites W189742998 @default.
• W2950646475 cites W1899621042 @default.
• W2950646475 cites W1967167074 @default.
• W2950646475 cites W1967306940 @default.
• W2950646475 cites W1968120338 @default.
• W2950646475 cites W197159984 @default.
• W2950646475 cites W1973194216 @default.
• W2950646475 cites W1973435495 @default.
• W2950646475 cites W1979859909 @default.
• W2950646475 cites W1985554184 @default.
• W2950646475 cites W1986280275 @default.
• W2950646475 cites W1992254844 @default.
• W2950646475 cites W1994807229 @default.
• W2950646475 cites W1995190797 @default.
• W2950646475 cites W2002658919 @default.
• W2950646475 cites W2020925091 @default.
• W2950646475 cites W2034368206 @default.
• W2950646475 cites W2037430361 @default.
• W2950646475 cites W2063513989 @default.
• W2950646475 cites W2064630832 @default.
• W2950646475 cites W2085536277 @default.
• W2950646475 cites W2087347434 @default.
• W2950646475 cites W2096536283 @default.
• W2950646475 cites W2098263691 @default.
• W2950646475 cites W2102689555 @default.
• W2950646475 cites W2103105543 @default.
• W2950646475 cites W2104267382 @default.
• W2950646475 cites W2107008379 @default.
• W2950646475 cites W2107861471 @default.
• W2950646475 cites W2108807072 @default.
• W2950646475 cites W2109181216 @default.
• W2950646475 cites W2111362445 @default.
• W2950646475 cites W2113369973 @default.
• W2950646475 cites W2114531539 @default.
• W2950646475 cites W2117525326 @default.
• W2950646475 cites W2119821739 @default.
• W2950646475 cites W2122922389 @default.
• W2950646475 cites W2123737232 @default.
• W2950646475 cites W2125398996 @default.
• W2950646475 cites W2129245267 @default.
• W2950646475 cites W2130698119 @default.
• W2950646475 cites W2130765257 @default.
• W2950646475 cites W2130996889 @default.
• W2950646475 cites W2133491790 @default.
• W2950646475 cites W2133958955 @default.
• W2950646475 cites W2135046866 @default.
• W2950646475 cites W2141840626 @default.
• W2950646475 cites W2143628714 @default.
• W2950646475 cites W2148418853 @default.
• W2950646475 cites W2149764047 @default.
• W2950646475 cites W2151040995 @default.
• W2950646475 cites W2153635508 @default.
• W2950646475 cites W2155398148 @default.
• W2950646475 cites W2156909104 @default.
• W2950646475 cites W2162071708 @default.
• W2950646475 cites W2165532891 @default.
• W2950646475 cites W2167250202 @default.
• W2950646475 cites W2172073485 @default.
• W2950646475 cites W2174647409 @default.
• W2950646475 cites W2296319761 @default.
• W2950646475 cites W246887631 @default.
• W2950646475 cites W2798909945 @default.
• W2950646475 cites W2988119488 @default.
• W2950646475 cites W740415 @default.
• W2950646475 cites W1589659394 @default.
• W2950646475 doi "https://doi.org/10.48550/arxiv.1009.4791" @default.
• W2950646475 hasPublicationYear "2010" @default.
• W2950646475 type Work @default.
• W2950646475 sameAs 2950646475 @default.
• W2950646475 citedByCount "0" @default.
• W2950646475 crossrefType "posted-content" @default.
• W2950646475 hasAuthorship W2950646475A5003668143 @default.
• W2950646475 hasAuthorship W2950646475A5042930568 @default.
• W2950646475 hasAuthorship W2950646475A5072744508 @default.
• W2950646475 hasAuthorship W2950646475A5081482638 @default.
• W2950646475 hasBestOaLocation W29506464751 @default.
• W2950646475 hasConcept C105795698 @default.
• W2950646475 hasConcept C11413529 @default.
• W2950646475 hasConcept C117251300 @default.
• W2950646475 hasConcept C12267149 @default.
• W2950646475 hasConcept C154945302 @default.
• W2950646475 hasConcept C199360897 @default.
• W2950646475 hasConcept C202444582 @default.
• W2950646475 hasConcept C2777735758 @default.
• W2950646475 hasConcept C33923547 @default.

• next