FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2087789467> ?p ?o ?g. }

• W2087789467 endingPage "5735" @default.
• W2087789467 startingPage "5716" @default.
• W2087789467 abstract "In this paper, an online learning algorithm is proposed as sequential stochastic approximation of a regularization path converging to the regression function in reproducing kernel Hilbert spaces (RKHSs). We show that it is possible to produce the best known strong (RKHS norm) convergence rate of batch learning, through a careful choice of the gain or step size sequences, depending on regularity assumptions on the regression function. The corresponding weak (mean square distance) convergence rate is optimal in the sense that it reaches the minimax and individual lower rates in the literature. In both cases we deduce almost sure convergence, using Bernstein-type inequalities for martingales in Hilbert spaces. To achieve this we develop a bias-variance decomposition similar to the batch learning setting; the bias consists in the approximation and drift errors along the regularization path, which display the same rates of convergence, and the variance arises from the sample error analysed as a (reverse) martingale difference sequence. The rates above are obtained by an optimal trade-off between the bias and the variance." @default.
• W2087789467 created "2016-06-24" @default.
• W2087789467 creator A5051299066 @default.
• W2087789467 creator A5054758400 @default.
• W2087789467 date "2014-09-01" @default.
• W2087789467 modified "2023-10-17" @default.
• W2087789467 title "Online Learning as Stochastic Approximation of Regularization Paths: Optimality and Almost-Sure Convergence" @default.
• W2087789467 cites W1484867920 @default.
• W2087789467 cites W1491706803 @default.
• W2087789467 cites W1510073064 @default.
• W2087789467 cites W1565176583 @default.
• W2087789467 cites W1570089119 @default.
• W2087789467 cites W1970781863 @default.
• W2087789467 cites W1986280275 @default.
• W2087789467 cites W1994616650 @default.
• W2087789467 cites W1996652862 @default.
• W2087789467 cites W2000021956 @default.
• W2087789467 cites W2009797711 @default.
• W2087789467 cites W2012501405 @default.
• W2087789467 cites W2022829242 @default.
• W2087789467 cites W2047311948 @default.
• W2087789467 cites W2052398345 @default.
• W2087789467 cites W2057682568 @default.
• W2087789467 cites W2059571389 @default.
• W2087789467 cites W2063978378 @default.
• W2087789467 cites W2073750241 @default.
• W2087789467 cites W2096840748 @default.
• W2087789467 cites W2108280195 @default.
• W2087789467 cites W2110652811 @default.
• W2087789467 cites W2118285149 @default.
• W2087789467 cites W2142563104 @default.
• W2087789467 cites W2150621701 @default.
• W2087789467 cites W2165867509 @default.
• W2087789467 cites W4230058301 @default.
• W2087789467 cites W4255272544 @default.
• W2087789467 cites W4298876635 @default.
• W2087789467 doi "https://doi.org/10.1109/tit.2014.2332531" @default.
• W2087789467 hasPublicationYear "2014" @default.
• W2087789467 type Work @default.
• W2087789467 sameAs 2087789467 @default.
• W2087789467 citedByCount "50" @default.
• W2087789467 countsByYear W20877894672014 @default.
• W2087789467 countsByYear W20877894672015 @default.
• W2087789467 countsByYear W20877894672016 @default.
• W2087789467 countsByYear W20877894672017 @default.
• W2087789467 countsByYear W20877894672018 @default.
• W2087789467 countsByYear W20877894672019 @default.
• W2087789467 countsByYear W20877894672020 @default.
• W2087789467 countsByYear W20877894672021 @default.
• W2087789467 countsByYear W20877894672022 @default.
• W2087789467 crossrefType "journal-article" @default.
• W2087789467 hasAuthorship W2087789467A5051299066 @default.
• W2087789467 hasAuthorship W2087789467A5054758400 @default.
• W2087789467 hasConcept C105795698 @default.
• W2087789467 hasConcept C122123141 @default.
• W2087789467 hasConcept C126255220 @default.
• W2087789467 hasConcept C148764684 @default.
• W2087789467 hasConcept C151319957 @default.
• W2087789467 hasConcept C154945302 @default.
• W2087789467 hasConcept C162324750 @default.
• W2087789467 hasConcept C194387892 @default.
• W2087789467 hasConcept C2776135515 @default.
• W2087789467 hasConcept C2777303404 @default.
• W2087789467 hasConcept C28826006 @default.
• W2087789467 hasConcept C31258907 @default.
• W2087789467 hasConcept C33923547 @default.
• W2087789467 hasConcept C41008148 @default.
• W2087789467 hasConcept C50522688 @default.
• W2087789467 hasConcept C55479107 @default.
• W2087789467 hasConcept C8272713 @default.
• W2087789467 hasConcept C95763700 @default.
• W2087789467 hasConceptScore W2087789467C105795698 @default.
• W2087789467 hasConceptScore W2087789467C122123141 @default.
• W2087789467 hasConceptScore W2087789467C126255220 @default.
• W2087789467 hasConceptScore W2087789467C148764684 @default.
• W2087789467 hasConceptScore W2087789467C151319957 @default.
• W2087789467 hasConceptScore W2087789467C154945302 @default.
• W2087789467 hasConceptScore W2087789467C162324750 @default.
• W2087789467 hasConceptScore W2087789467C194387892 @default.
• W2087789467 hasConceptScore W2087789467C2776135515 @default.
• W2087789467 hasConceptScore W2087789467C2777303404 @default.
• W2087789467 hasConceptScore W2087789467C28826006 @default.
• W2087789467 hasConceptScore W2087789467C31258907 @default.
• W2087789467 hasConceptScore W2087789467C33923547 @default.
• W2087789467 hasConceptScore W2087789467C41008148 @default.
• W2087789467 hasConceptScore W2087789467C50522688 @default.
• W2087789467 hasConceptScore W2087789467C55479107 @default.
• W2087789467 hasConceptScore W2087789467C8272713 @default.
• W2087789467 hasConceptScore W2087789467C95763700 @default.
• W2087789467 hasFunder F4320321001 @default.
• W2087789467 hasFunder F4320335777 @default.
• W2087789467 hasIssue "9" @default.
• W2087789467 hasLocation W20877894671 @default.
• W2087789467 hasOpenAccess W2087789467 @default.
• W2087789467 hasPrimaryLocation W20877894671 @default.
• W2087789467 hasRelatedWork W1978379809 @default.
• W2087789467 hasRelatedWork W1981220283 @default.
• W2087789467 hasRelatedWork W1999620897 @default.

• next