Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225557543> ?p ?o ?g. }
- W4225557543 endingPage "7573" @default.
- W4225557543 startingPage "7561" @default.
- W4225557543 abstract "We propose a novel semi-supervised learning (SSL) assumption—the high separation probability (HSP) assumption—that is complementary to the common low density separation (LDS) assumption, and a highly accurate and efficient SSL algorithm, the transductive minimax probability machine (TMPM). Under the HSP assumption, a preferred classifier should ensure that with high probability the samples are far away from the decision boundary. Similarly, the LDS assumption states that the decision boundary should lie in a low-density region. A remarkable property of HSP is that it provides an elegant theoretical framework for analyzing the generalization performance of SSL, while it remains difficult for the LDS assumption to have a quantitative criterion on how it fits the data. Moreover, it has the same paradigm as the minimax probability machine (MPM) and yields an intractable mixed-integer optimization for solving the transductive label assignment problem. Consequently, we propose TMPM with a label-switching strategy and iterative manner to formalize our approximate solution. Theoretically, we analyze the computational complexity, convergence property, and generalization error bound of TMPM. Empirically, we show that TMPM achieves competitive performance on a wide range of datasets, while being almost one order of magnitude faster than existing SSL approaches." @default.
- W4225557543 created "2022-05-05" @default.
- W4225557543 creator A5013240918 @default.
- W4225557543 creator A5062340204 @default.
- W4225557543 date "2022-12-01" @default.
- W4225557543 modified "2023-09-27" @default.
- W4225557543 title "The High Separation Probability Assumption for Semi-Supervised Learning" @default.
- W4225557543 cites W1433051557 @default.
- W4225557543 cites W1479807131 @default.
- W4225557543 cites W1517228427 @default.
- W4225557543 cites W1807914171 @default.
- W4225557543 cites W1968817125 @default.
- W4225557543 cites W1969198379 @default.
- W4225557543 cites W1981613567 @default.
- W4225557543 cites W1999478721 @default.
- W4225557543 cites W2009923109 @default.
- W4225557543 cites W2012844989 @default.
- W4225557543 cites W2019288156 @default.
- W4225557543 cites W2044195942 @default.
- W4225557543 cites W2048679005 @default.
- W4225557543 cites W2073614810 @default.
- W4225557543 cites W2079057609 @default.
- W4225557543 cites W2083885588 @default.
- W4225557543 cites W2097089247 @default.
- W4225557543 cites W2106401878 @default.
- W4225557543 cites W2107074288 @default.
- W4225557543 cites W2113971713 @default.
- W4225557543 cites W2128097790 @default.
- W4225557543 cites W2145376937 @default.
- W4225557543 cites W2148029428 @default.
- W4225557543 cites W2154546490 @default.
- W4225557543 cites W2203392907 @default.
- W4225557543 cites W2260382512 @default.
- W4225557543 cites W2398821504 @default.
- W4225557543 cites W2563949017 @default.
- W4225557543 cites W2594323454 @default.
- W4225557543 cites W2731977498 @default.
- W4225557543 cites W2763583057 @default.
- W4225557543 cites W2782698363 @default.
- W4225557543 cites W2784620354 @default.
- W4225557543 cites W2890740717 @default.
- W4225557543 cites W2896370244 @default.
- W4225557543 cites W2913252837 @default.
- W4225557543 cites W2928823069 @default.
- W4225557543 cites W2950672152 @default.
- W4225557543 cites W2963080758 @default.
- W4225557543 cites W2963558289 @default.
- W4225557543 cites W2964159205 @default.
- W4225557543 cites W2966415767 @default.
- W4225557543 cites W2969974905 @default.
- W4225557543 cites W3015997807 @default.
- W4225557543 cites W3133588596 @default.
- W4225557543 cites W38891395 @default.
- W4225557543 doi "https://doi.org/10.1109/tsmc.2022.3161067" @default.
- W4225557543 hasPublicationYear "2022" @default.
- W4225557543 type Work @default.
- W4225557543 citedByCount "2" @default.
- W4225557543 countsByYear W42255575432023 @default.
- W4225557543 crossrefType "journal-article" @default.
- W4225557543 hasAuthorship W4225557543A5013240918 @default.
- W4225557543 hasAuthorship W4225557543A5062340204 @default.
- W4225557543 hasConcept C111472728 @default.
- W4225557543 hasConcept C11413529 @default.
- W4225557543 hasConcept C119857082 @default.
- W4225557543 hasConcept C126255220 @default.
- W4225557543 hasConcept C134306372 @default.
- W4225557543 hasConcept C138885662 @default.
- W4225557543 hasConcept C149728462 @default.
- W4225557543 hasConcept C154945302 @default.
- W4225557543 hasConcept C177148314 @default.
- W4225557543 hasConcept C189950617 @default.
- W4225557543 hasConcept C2778445095 @default.
- W4225557543 hasConcept C33923547 @default.
- W4225557543 hasConcept C41008148 @default.
- W4225557543 hasConcept C42023084 @default.
- W4225557543 hasConcept C62354387 @default.
- W4225557543 hasConcept C95623464 @default.
- W4225557543 hasConceptScore W4225557543C111472728 @default.
- W4225557543 hasConceptScore W4225557543C11413529 @default.
- W4225557543 hasConceptScore W4225557543C119857082 @default.
- W4225557543 hasConceptScore W4225557543C126255220 @default.
- W4225557543 hasConceptScore W4225557543C134306372 @default.
- W4225557543 hasConceptScore W4225557543C138885662 @default.
- W4225557543 hasConceptScore W4225557543C149728462 @default.
- W4225557543 hasConceptScore W4225557543C154945302 @default.
- W4225557543 hasConceptScore W4225557543C177148314 @default.
- W4225557543 hasConceptScore W4225557543C189950617 @default.
- W4225557543 hasConceptScore W4225557543C2778445095 @default.
- W4225557543 hasConceptScore W4225557543C33923547 @default.
- W4225557543 hasConceptScore W4225557543C41008148 @default.
- W4225557543 hasConceptScore W4225557543C42023084 @default.
- W4225557543 hasConceptScore W4225557543C62354387 @default.
- W4225557543 hasConceptScore W4225557543C95623464 @default.
- W4225557543 hasFunder F4320321001 @default.
- W4225557543 hasFunder F4320335777 @default.
- W4225557543 hasIssue "12" @default.
- W4225557543 hasLocation W42255575431 @default.
- W4225557543 hasOpenAccess W4225557543 @default.