FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2901020299> ?p ?o ?g. }

• W2901020299 endingPage "2092" @default.
• W2901020299 startingPage "2079" @default.
• W2901020299 abstract "In this paper, we investigate the problem of sparse online linear classification in changing environments. We first analyze the tracking performance of standard online linear classifiers, which use gradient descent for minimizing the regularized hinge loss. The derived shifting bounds highlight the importance of choosing appropriate step sizes in the presence of concept drifts. Notably, we show that a better adaptability to concept drifts can be achieved using constant step sizes rather than the state-of-the-art decreasing step sizes. Based on these observations, we then propose a novel sparse approximated linear classifier, called sparse approximated linear classification (SALC), which uses a constant step size. In essence, SALC simply rounds small weights to zero for achieving sparsity and controls the truncation error in a principled way for achieving a low tracking regret. The degree of sparsity obtained by SALC is continuous and can be controlled by a parameter which captures the tradeoff between the sparsity of the model and the regret performance of the algorithm. Experiments on nine stationary data sets show that SALC is superior to the state-of-the-art sparse online learning algorithms, especially when the solution is required to be sparse; on seven groups of nonstationary data sets with various total shifting amounts, SALC also presents a good ability to track drifts. When wrapped with a drift detector, SALC achieves a remarkable tracking performance regardless of the total shifting amount." @default.
• W2901020299 created "2018-11-29" @default.
• W2901020299 creator A5000297333 @default.
• W2901020299 creator A5041175408 @default.
• W2901020299 creator A5074250521 @default.
• W2901020299 creator A5075449658 @default.
• W2901020299 date "2019-07-01" @default.
• W2901020299 modified "2023-10-16" @default.
• W2901020299 title "Tracking Sparse Linear Classifiers" @default.
• W2901020299 cites W1585854823 @default.
• W2901020299 cites W1973678258 @default.
• W2901020299 cites W1978259121 @default.
• W2901020299 cites W1982039810 @default.
• W2901020299 cites W1991076541 @default.
• W2901020299 cites W1994520254 @default.
• W2901020299 cites W2000454347 @default.
• W2901020299 cites W2021739900 @default.
• W2901020299 cites W2028781966 @default.
• W2901020299 cites W2028870622 @default.
• W2901020299 cites W2091825929 @default.
• W2901020299 cites W2099419573 @default.
• W2901020299 cites W2122537498 @default.
• W2901020299 cites W2125993116 @default.
• W2901020299 cites W2136051823 @default.
• W2901020299 cites W2146571341 @default.
• W2901020299 cites W2149882723 @default.
• W2901020299 cites W2150621701 @default.
• W2901020299 cites W2180180824 @default.
• W2901020299 cites W2209420799 @default.
• W2901020299 cites W2585528949 @default.
• W2901020299 cites W2600796512 @default.
• W2901020299 doi "https://doi.org/10.1109/tnnls.2018.2877433" @default.
• W2901020299 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30442619" @default.
• W2901020299 hasPublicationYear "2019" @default.
• W2901020299 type Work @default.
• W2901020299 sameAs 2901020299 @default.
• W2901020299 citedByCount "8" @default.
• W2901020299 countsByYear W29010202992019 @default.
• W2901020299 countsByYear W29010202992020 @default.
• W2901020299 countsByYear W29010202992021 @default.
• W2901020299 countsByYear W29010202992022 @default.
• W2901020299 countsByYear W29010202992023 @default.
• W2901020299 crossrefType "journal-article" @default.
• W2901020299 hasAuthorship W2901020299A5000297333 @default.
• W2901020299 hasAuthorship W2901020299A5041175408 @default.
• W2901020299 hasAuthorship W2901020299A5074250521 @default.
• W2901020299 hasAuthorship W2901020299A5075449658 @default.
• W2901020299 hasConcept C11413529 @default.
• W2901020299 hasConcept C119857082 @default.
• W2901020299 hasConcept C139532973 @default.
• W2901020299 hasConcept C153180895 @default.
• W2901020299 hasConcept C153258448 @default.
• W2901020299 hasConcept C154945302 @default.
• W2901020299 hasConcept C15744967 @default.
• W2901020299 hasConcept C19417346 @default.
• W2901020299 hasConcept C199360897 @default.
• W2901020299 hasConcept C2775936607 @default.
• W2901020299 hasConcept C2777027219 @default.
• W2901020299 hasConcept C33923547 @default.
• W2901020299 hasConcept C41008148 @default.
• W2901020299 hasConcept C50644808 @default.
• W2901020299 hasConcept C50817715 @default.
• W2901020299 hasConcept C95623464 @default.
• W2901020299 hasConceptScore W2901020299C11413529 @default.
• W2901020299 hasConceptScore W2901020299C119857082 @default.
• W2901020299 hasConceptScore W2901020299C139532973 @default.
• W2901020299 hasConceptScore W2901020299C153180895 @default.
• W2901020299 hasConceptScore W2901020299C153258448 @default.
• W2901020299 hasConceptScore W2901020299C154945302 @default.
• W2901020299 hasConceptScore W2901020299C15744967 @default.
• W2901020299 hasConceptScore W2901020299C19417346 @default.
• W2901020299 hasConceptScore W2901020299C199360897 @default.
• W2901020299 hasConceptScore W2901020299C2775936607 @default.
• W2901020299 hasConceptScore W2901020299C2777027219 @default.
• W2901020299 hasConceptScore W2901020299C33923547 @default.
• W2901020299 hasConceptScore W2901020299C41008148 @default.
• W2901020299 hasConceptScore W2901020299C50644808 @default.
• W2901020299 hasConceptScore W2901020299C50817715 @default.
• W2901020299 hasConceptScore W2901020299C95623464 @default.
• W2901020299 hasFunder F4320321001 @default.
• W2901020299 hasFunder F4320322769 @default.
• W2901020299 hasFunder F4320335777 @default.
• W2901020299 hasIssue "7" @default.
• W2901020299 hasLocation W29010202991 @default.
• W2901020299 hasLocation W29010202992 @default.
• W2901020299 hasOpenAccess W2901020299 @default.
• W2901020299 hasPrimaryLocation W29010202991 @default.
• W2901020299 hasRelatedWork W2001652754 @default.
• W2901020299 hasRelatedWork W2379065761 @default.
• W2901020299 hasRelatedWork W2549006548 @default.
• W2901020299 hasRelatedWork W2807311372 @default.
• W2901020299 hasRelatedWork W2921036759 @default.
• W2901020299 hasRelatedWork W2972035100 @default.
• W2901020299 hasRelatedWork W3043252291 @default.
• W2901020299 hasRelatedWork W4214932115 @default.
• W2901020299 hasRelatedWork W4221015625 @default.
• W2901020299 hasRelatedWork W3158004940 @default.
• W2901020299 hasVolume "30" @default.

• next