Matches in SemOpenAlex for { <https://semopenalex.org/work/W3172624703> ?p ?o ?g. }
- W3172624703 abstract "The support vector machine (SVM) and minimum Euclidean norm least squares regression are two fundamentally different approaches to fitting linear models, but they have recently been connected in models for very high-dimensional data through a phenomenon of support vector proliferation, where every training example used to fit an SVM becomes a support vector. In this paper, we explore the generality of this phenomenon and make the following contributions. First, we prove a super-linear lower bound on the dimension (in terms of sample size) required for support vector proliferation in independent feature models, matching the upper bounds from previous works. We further identify a sharp phase transition in Gaussian feature models, bound the width of this transition, and give experimental support for its universality. Finally, we hypothesize that this phase transition occurs only in much higher-dimensional settings in the $ell_1$ variant of the SVM, and we present a new geometric characterization of the problem that may elucidate this phenomenon for the general $ell_p$ case." @default.
- W3172624703 created "2021-06-22" @default.
- W3172624703 creator A5020375684 @default.
- W3172624703 creator A5061246300 @default.
- W3172624703 creator A5074033914 @default.
- W3172624703 date "2021-05-28" @default.
- W3172624703 modified "2023-09-27" @default.
- W3172624703 title "Support vector machines and linear regression coincide with very high-dimensional features" @default.
- W3172624703 cites W1530699444 @default.
- W3172624703 cites W1564584004 @default.
- W3172624703 cites W1573820523 @default.
- W3172624703 cites W1599545414 @default.
- W3172624703 cites W1605479404 @default.
- W3172624703 cites W2004915807 @default.
- W3172624703 cites W2020658897 @default.
- W3172624703 cites W2021045253 @default.
- W3172624703 cites W2025223969 @default.
- W3172624703 cites W2027040307 @default.
- W3172624703 cites W2089880431 @default.
- W3172624703 cites W2111049014 @default.
- W3172624703 cites W2115275122 @default.
- W3172624703 cites W2122456939 @default.
- W3172624703 cites W2135046866 @default.
- W3172624703 cites W2149175418 @default.
- W3172624703 cites W2161278885 @default.
- W3172624703 cites W2161920802 @default.
- W3172624703 cites W2294690908 @default.
- W3172624703 cites W2626872000 @default.
- W3172624703 cites W2911742574 @default.
- W3172624703 cites W2944156084 @default.
- W3172624703 cites W2948069823 @default.
- W3172624703 cites W2965772785 @default.
- W3172624703 cites W2967536008 @default.
- W3172624703 cites W2977400275 @default.
- W3172624703 cites W2998511956 @default.
- W3172624703 cites W3004510491 @default.
- W3172624703 cites W3005822671 @default.
- W3172624703 cites W3007237875 @default.
- W3172624703 cites W3014316192 @default.
- W3172624703 cites W3018252856 @default.
- W3172624703 cites W3107776541 @default.
- W3172624703 cites W3111350549 @default.
- W3172624703 cites W3151795292 @default.
- W3172624703 cites W3157216945 @default.
- W3172624703 cites W3157298807 @default.
- W3172624703 cites W3157730063 @default.
- W3172624703 cites W3157731489 @default.
- W3172624703 cites W3168158992 @default.
- W3172624703 cites W3204592394 @default.
- W3172624703 cites W744717813 @default.
- W3172624703 hasPublicationYear "2021" @default.
- W3172624703 type Work @default.
- W3172624703 sameAs 3172624703 @default.
- W3172624703 citedByCount "2" @default.
- W3172624703 countsByYear W31726247032021 @default.
- W3172624703 crossrefType "posted-content" @default.
- W3172624703 hasAuthorship W3172624703A5020375684 @default.
- W3172624703 hasAuthorship W3172624703A5061246300 @default.
- W3172624703 hasAuthorship W3172624703A5074033914 @default.
- W3172624703 hasConcept C11413529 @default.
- W3172624703 hasConcept C114614502 @default.
- W3172624703 hasConcept C120174047 @default.
- W3172624703 hasConcept C121332964 @default.
- W3172624703 hasConcept C12267149 @default.
- W3172624703 hasConcept C134306372 @default.
- W3172624703 hasConcept C145828037 @default.
- W3172624703 hasConcept C154945302 @default.
- W3172624703 hasConcept C15744967 @default.
- W3172624703 hasConcept C183992945 @default.
- W3172624703 hasConcept C2780767217 @default.
- W3172624703 hasConcept C33676613 @default.
- W3172624703 hasConcept C33923547 @default.
- W3172624703 hasConcept C41008148 @default.
- W3172624703 hasConcept C50335755 @default.
- W3172624703 hasConcept C542102704 @default.
- W3172624703 hasConcept C62520636 @default.
- W3172624703 hasConcept C77553402 @default.
- W3172624703 hasConceptScore W3172624703C11413529 @default.
- W3172624703 hasConceptScore W3172624703C114614502 @default.
- W3172624703 hasConceptScore W3172624703C120174047 @default.
- W3172624703 hasConceptScore W3172624703C121332964 @default.
- W3172624703 hasConceptScore W3172624703C12267149 @default.
- W3172624703 hasConceptScore W3172624703C134306372 @default.
- W3172624703 hasConceptScore W3172624703C145828037 @default.
- W3172624703 hasConceptScore W3172624703C154945302 @default.
- W3172624703 hasConceptScore W3172624703C15744967 @default.
- W3172624703 hasConceptScore W3172624703C183992945 @default.
- W3172624703 hasConceptScore W3172624703C2780767217 @default.
- W3172624703 hasConceptScore W3172624703C33676613 @default.
- W3172624703 hasConceptScore W3172624703C33923547 @default.
- W3172624703 hasConceptScore W3172624703C41008148 @default.
- W3172624703 hasConceptScore W3172624703C50335755 @default.
- W3172624703 hasConceptScore W3172624703C542102704 @default.
- W3172624703 hasConceptScore W3172624703C62520636 @default.
- W3172624703 hasConceptScore W3172624703C77553402 @default.
- W3172624703 hasLocation W31726247031 @default.
- W3172624703 hasOpenAccess W3172624703 @default.
- W3172624703 hasPrimaryLocation W31726247031 @default.
- W3172624703 hasRelatedWork W1489577094 @default.
- W3172624703 hasRelatedWork W1495350765 @default.