FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2156005874> ?p ?o ?g. }

• W2156005874 endingPage "1241" @default.
• W2156005874 startingPage "1233" @default.
• W2156005874 abstract "In large scale learning problems it is often easy to collect simple statistics of the data, but hard or impractical to store all the original data. A key question in this setting is how to construct classifiers based on such partial information. One traditional approach to the problem has been to use maximum entropy arguments to induce a complete distribution on variables from statistics. However, this approach essentially makes conditional independence assumptions about the distribution, and furthermore does not optimize prediction loss. Here we present a framework for discriminative learning given a set of statistics. Specifically, we address the case where all variables are discrete and we have access to various marginals. Our approach minimizes the worst case hinge loss in this case, which upper bounds the generalization error. We show that for certain sets of statistics the problem is tractable, and in the general case can be approximated using MAP LP relaxations. Empirical results show that the method is competitive with other approaches that use the same input." @default.
• W2156005874 created "2016-06-24" @default.
• W2156005874 creator A5043019006 @default.
• W2156005874 creator A5044999818 @default.
• W2156005874 creator A5047817959 @default.
• W2156005874 date "2014-06-21" @default.
• W2156005874 modified "2023-09-25" @default.
• W2156005874 title "Discrete Chebyshev Classifiers" @default.
• W2156005874 cites W1511986666 @default.
• W2156005874 cites W1517228427 @default.
• W2156005874 cites W1517526469 @default.
• W2156005874 cites W1590693676 @default.
• W2156005874 cites W1817561967 @default.
• W2156005874 cites W1972284617 @default.
• W2156005874 cites W1988520084 @default.
• W2156005874 cites W1995897489 @default.
• W2156005874 cites W2061293389 @default.
• W2156005874 cites W2074184884 @default.
• W2156005874 cites W2089105401 @default.
• W2156005874 cites W2100556411 @default.
• W2156005874 cites W2100726809 @default.
• W2156005874 cites W2107467870 @default.
• W2156005874 cites W2107961375 @default.
• W2156005874 cites W2120340025 @default.
• W2156005874 cites W2128884519 @default.
• W2156005874 cites W2131538180 @default.
• W2156005874 cites W2135414191 @default.
• W2156005874 cites W2149474573 @default.
• W2156005874 cites W2153417333 @default.
• W2156005874 cites W2157791002 @default.
• W2156005874 cites W2158231134 @default.
• W2156005874 cites W2167732364 @default.
• W2156005874 cites W2525721411 @default.
• W2156005874 cites W2798766386 @default.
• W2156005874 cites W2964243709 @default.
• W2156005874 cites W2974222084 @default.
• W2156005874 cites W2107288260 @default.
• W2156005874 hasPublicationYear "2014" @default.
• W2156005874 type Work @default.
• W2156005874 sameAs 2156005874 @default.
• W2156005874 citedByCount "9" @default.
• W2156005874 countsByYear W21560058742014 @default.
• W2156005874 countsByYear W21560058742015 @default.
• W2156005874 countsByYear W21560058742016 @default.
• W2156005874 countsByYear W21560058742019 @default.
• W2156005874 crossrefType "proceedings-article" @default.
• W2156005874 hasAuthorship W2156005874A5043019006 @default.
• W2156005874 hasAuthorship W2156005874A5044999818 @default.
• W2156005874 hasAuthorship W2156005874A5047817959 @default.
• W2156005874 hasConcept C105795698 @default.
• W2156005874 hasConcept C106301342 @default.
• W2156005874 hasConcept C121332964 @default.
• W2156005874 hasConcept C12267149 @default.
• W2156005874 hasConcept C126255220 @default.
• W2156005874 hasConcept C134306372 @default.
• W2156005874 hasConcept C149441793 @default.
• W2156005874 hasConcept C154945302 @default.
• W2156005874 hasConcept C177148314 @default.
• W2156005874 hasConcept C33923547 @default.
• W2156005874 hasConcept C35651441 @default.
• W2156005874 hasConcept C39891107 @default.
• W2156005874 hasConcept C41008148 @default.
• W2156005874 hasConcept C62520636 @default.
• W2156005874 hasConcept C97931131 @default.
• W2156005874 hasConceptScore W2156005874C105795698 @default.
• W2156005874 hasConceptScore W2156005874C106301342 @default.
• W2156005874 hasConceptScore W2156005874C121332964 @default.
• W2156005874 hasConceptScore W2156005874C12267149 @default.
• W2156005874 hasConceptScore W2156005874C126255220 @default.
• W2156005874 hasConceptScore W2156005874C134306372 @default.
• W2156005874 hasConceptScore W2156005874C149441793 @default.
• W2156005874 hasConceptScore W2156005874C154945302 @default.
• W2156005874 hasConceptScore W2156005874C177148314 @default.
• W2156005874 hasConceptScore W2156005874C33923547 @default.
• W2156005874 hasConceptScore W2156005874C35651441 @default.
• W2156005874 hasConceptScore W2156005874C39891107 @default.
• W2156005874 hasConceptScore W2156005874C41008148 @default.
• W2156005874 hasConceptScore W2156005874C62520636 @default.
• W2156005874 hasConceptScore W2156005874C97931131 @default.
• W2156005874 hasLocation W21560058741 @default.
• W2156005874 hasOpenAccess W2156005874 @default.
• W2156005874 hasPrimaryLocation W21560058741 @default.
• W2156005874 hasRelatedWork W15601695 @default.
• W2156005874 hasRelatedWork W1904555434 @default.
• W2156005874 hasRelatedWork W1914062059 @default.
• W2156005874 hasRelatedWork W2107961375 @default.
• W2156005874 hasRelatedWork W2135036692 @default.
• W2156005874 hasRelatedWork W24455790 @default.
• W2156005874 hasRelatedWork W28886848 @default.
• W2156005874 hasRelatedWork W2896214533 @default.
• W2156005874 hasRelatedWork W2950906236 @default.
• W2156005874 hasRelatedWork W2964243709 @default.
• W2156005874 hasRelatedWork W3005529192 @default.
• W2156005874 hasRelatedWork W3033614773 @default.
• W2156005874 hasRelatedWork W3091852560 @default.
• W2156005874 hasRelatedWork W3094500068 @default.
• W2156005874 hasRelatedWork W3110394586 @default.
• W2156005874 hasRelatedWork W3128171061 @default.

• next