Matches in SemOpenAlex for { <https://semopenalex.org/work/W2808062556> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2808062556 abstract "In contrast to offline kernel selection, online kernel selection must rise to the new challenges of passing the training set once, selecting optimal kernels and updating hypotheses at each round, enjoying a sublinear regret bound for online kernel learning, and requiring a constant maintenance time complexity at each round and an efficient overall time complexity integrated with online kernel learning. However, most of existing online kernel selection approaches can not meet the new challenges. To address this issue, we propose a novel online kernel selection approach via the incremental sketched kernel alignment criterion, which meets all the new challenges. We first define the incremental sketched kernel alignment (ISKA) criterion, which estimates the kernel alignment and can be computed incrementally and efficiently. When applying the proposed ISKA criterion to online kernel selection, we adopt the subclass coherence to maintain the hypothesis space, select the optimal kernel at each round using the median of the ISKA criterion estimates, and update the hypothesis following the online gradient decent method. We prove that the ISKA criterion is an unbiased estimate of the maximum mean discrepancy, enjoys the optimal logarithmic regret bound for online kernel learning, and has a constant maintenance time complexity at each round and a logarithmic overall time complexity integrated with online kernel learning. Empirical studies demonstrate that the proposed online kernel selection approach is computationally efficient while maintaining comparable accuracy for online kernel learning." @default.
- W2808062556 created "2018-06-21" @default.
- W2808062556 creator A5047065510 @default.
- W2808062556 creator A5089843351 @default.
- W2808062556 date "2018-07-01" @default.
- W2808062556 modified "2023-09-24" @default.
- W2808062556 title "Online Kernel Selection via Incremental Sketched Kernel Alignment" @default.
- W2808062556 cites W1495978835 @default.
- W2808062556 cites W1568179240 @default.
- W2808062556 cites W1837744813 @default.
- W2808062556 cites W2005447957 @default.
- W2808062556 cites W2012501405 @default.
- W2808062556 cites W2098102888 @default.
- W2808062556 cites W2116148865 @default.
- W2808062556 cites W2119885577 @default.
- W2808062556 cites W2121038136 @default.
- W2808062556 cites W2125335195 @default.
- W2808062556 cites W2150621701 @default.
- W2808062556 cites W2549898883 @default.
- W2808062556 cites W2579923771 @default.
- W2808062556 cites W2741352581 @default.
- W2808062556 doi "https://doi.org/10.24963/ijcai.2018/433" @default.
- W2808062556 hasPublicationYear "2018" @default.
- W2808062556 type Work @default.
- W2808062556 sameAs 2808062556 @default.
- W2808062556 citedByCount "10" @default.
- W2808062556 countsByYear W28080625562018 @default.
- W2808062556 countsByYear W28080625562019 @default.
- W2808062556 countsByYear W28080625562020 @default.
- W2808062556 countsByYear W28080625562021 @default.
- W2808062556 countsByYear W28080625562022 @default.
- W2808062556 countsByYear W28080625562023 @default.
- W2808062556 crossrefType "proceedings-article" @default.
- W2808062556 hasAuthorship W2808062556A5047065510 @default.
- W2808062556 hasAuthorship W2808062556A5089843351 @default.
- W2808062556 hasBestOaLocation W28080625561 @default.
- W2808062556 hasConcept C118615104 @default.
- W2808062556 hasConcept C119857082 @default.
- W2808062556 hasConcept C122280245 @default.
- W2808062556 hasConcept C12267149 @default.
- W2808062556 hasConcept C126255220 @default.
- W2808062556 hasConcept C134517425 @default.
- W2808062556 hasConcept C154945302 @default.
- W2808062556 hasConcept C195699287 @default.
- W2808062556 hasConcept C27406209 @default.
- W2808062556 hasConcept C33923547 @default.
- W2808062556 hasConcept C41008148 @default.
- W2808062556 hasConcept C50817715 @default.
- W2808062556 hasConcept C74193536 @default.
- W2808062556 hasConcept C75866337 @default.
- W2808062556 hasConcept C81917197 @default.
- W2808062556 hasConceptScore W2808062556C118615104 @default.
- W2808062556 hasConceptScore W2808062556C119857082 @default.
- W2808062556 hasConceptScore W2808062556C122280245 @default.
- W2808062556 hasConceptScore W2808062556C12267149 @default.
- W2808062556 hasConceptScore W2808062556C126255220 @default.
- W2808062556 hasConceptScore W2808062556C134517425 @default.
- W2808062556 hasConceptScore W2808062556C154945302 @default.
- W2808062556 hasConceptScore W2808062556C195699287 @default.
- W2808062556 hasConceptScore W2808062556C27406209 @default.
- W2808062556 hasConceptScore W2808062556C33923547 @default.
- W2808062556 hasConceptScore W2808062556C41008148 @default.
- W2808062556 hasConceptScore W2808062556C50817715 @default.
- W2808062556 hasConceptScore W2808062556C74193536 @default.
- W2808062556 hasConceptScore W2808062556C75866337 @default.
- W2808062556 hasConceptScore W2808062556C81917197 @default.
- W2808062556 hasLocation W28080625561 @default.
- W2808062556 hasOpenAccess W2808062556 @default.
- W2808062556 hasPrimaryLocation W28080625561 @default.
- W2808062556 hasRelatedWork W1603091392 @default.
- W2808062556 hasRelatedWork W1984421104 @default.
- W2808062556 hasRelatedWork W2079825755 @default.
- W2808062556 hasRelatedWork W2095626363 @default.
- W2808062556 hasRelatedWork W2367778514 @default.
- W2808062556 hasRelatedWork W2371064519 @default.
- W2808062556 hasRelatedWork W2382704364 @default.
- W2808062556 hasRelatedWork W2420851703 @default.
- W2808062556 hasRelatedWork W2534603691 @default.
- W2808062556 hasRelatedWork W4311138679 @default.
- W2808062556 isParatext "false" @default.
- W2808062556 isRetracted "false" @default.
- W2808062556 magId "2808062556" @default.
- W2808062556 workType "article" @default.