Matches in SemOpenAlex for { <https://semopenalex.org/work/W3006500171> ?p ?o ?g. }
- W3006500171 abstract "We study kernel selection for online kernel learning, also known as online kernel selection which can be treated as a sequential decision problem and thus must balance the regret and the time complexity. Existing online kernel selection approaches via expert advice and classical adversarial bandit model can not meet the issue. In this work, we propose a novel grouped adversarial bandit solution to the problem. We first correspond each candidate kernel to a basic arm of an adversarial bandit problem. Then, all of the kernels are divided into several groups where each group is abstracted as a super arm. At each round, we choose a super arm and a basic kernel within the selected super arm, and make prediction by an online kernel learning algorithm. Besides, we introduce a Bernoulli random variable to decide whether to choose all of the rest super arms. Theoretical analysis shows the proposed approach balances the regret and the time complexity explicitly, which could enjoy better pseudo-regret and high probability regret bound than classical adversarial bandit model and lighter time complexity than expert advice model. Experimental results on benchmark datasets verify that the proposed approach balances the efficiency and effectiveness better." @default.
- W3006500171 created "2020-02-24" @default.
- W3006500171 creator A5083664091 @default.
- W3006500171 creator A5089843351 @default.
- W3006500171 date "2019-11-01" @default.
- W3006500171 modified "2023-09-26" @default.
- W3006500171 title "Online Kernel Selection via Grouped Adversarial Bandit Model" @default.
- W3006500171 cites W1492499319 @default.
- W3006500171 cites W1498977598 @default.
- W3006500171 cites W1510812122 @default.
- W3006500171 cites W1570963478 @default.
- W3006500171 cites W1788769502 @default.
- W3006500171 cites W2049934117 @default.
- W3006500171 cites W2051645349 @default.
- W3006500171 cites W2077723394 @default.
- W3006500171 cites W2101379030 @default.
- W3006500171 cites W2125335195 @default.
- W3006500171 cites W2132087961 @default.
- W3006500171 cites W2152951864 @default.
- W3006500171 cites W2161603851 @default.
- W3006500171 cites W2161704742 @default.
- W3006500171 cites W2213469232 @default.
- W3006500171 cites W2408432900 @default.
- W3006500171 cites W2739765643 @default.
- W3006500171 cites W2741352581 @default.
- W3006500171 cites W2796280559 @default.
- W3006500171 cites W2808062556 @default.
- W3006500171 cites W2886019888 @default.
- W3006500171 cites W2946399799 @default.
- W3006500171 cites W2963137828 @default.
- W3006500171 cites W2963389158 @default.
- W3006500171 doi "https://doi.org/10.1109/ictai.2019.00100" @default.
- W3006500171 hasPublicationYear "2019" @default.
- W3006500171 type Work @default.
- W3006500171 sameAs 3006500171 @default.
- W3006500171 citedByCount "0" @default.
- W3006500171 crossrefType "proceedings-article" @default.
- W3006500171 hasAuthorship W3006500171A5083664091 @default.
- W3006500171 hasAuthorship W3006500171A5089843351 @default.
- W3006500171 hasConcept C114614502 @default.
- W3006500171 hasConcept C119857082 @default.
- W3006500171 hasConcept C122280245 @default.
- W3006500171 hasConcept C12267149 @default.
- W3006500171 hasConcept C123197309 @default.
- W3006500171 hasConcept C126255220 @default.
- W3006500171 hasConcept C13280743 @default.
- W3006500171 hasConcept C154945302 @default.
- W3006500171 hasConcept C185798385 @default.
- W3006500171 hasConcept C205649164 @default.
- W3006500171 hasConcept C2776879701 @default.
- W3006500171 hasConcept C33923547 @default.
- W3006500171 hasConcept C37736160 @default.
- W3006500171 hasConcept C41008148 @default.
- W3006500171 hasConcept C50817715 @default.
- W3006500171 hasConcept C73602740 @default.
- W3006500171 hasConcept C74193536 @default.
- W3006500171 hasConcept C81917197 @default.
- W3006500171 hasConceptScore W3006500171C114614502 @default.
- W3006500171 hasConceptScore W3006500171C119857082 @default.
- W3006500171 hasConceptScore W3006500171C122280245 @default.
- W3006500171 hasConceptScore W3006500171C12267149 @default.
- W3006500171 hasConceptScore W3006500171C123197309 @default.
- W3006500171 hasConceptScore W3006500171C126255220 @default.
- W3006500171 hasConceptScore W3006500171C13280743 @default.
- W3006500171 hasConceptScore W3006500171C154945302 @default.
- W3006500171 hasConceptScore W3006500171C185798385 @default.
- W3006500171 hasConceptScore W3006500171C205649164 @default.
- W3006500171 hasConceptScore W3006500171C2776879701 @default.
- W3006500171 hasConceptScore W3006500171C33923547 @default.
- W3006500171 hasConceptScore W3006500171C37736160 @default.
- W3006500171 hasConceptScore W3006500171C41008148 @default.
- W3006500171 hasConceptScore W3006500171C50817715 @default.
- W3006500171 hasConceptScore W3006500171C73602740 @default.
- W3006500171 hasConceptScore W3006500171C74193536 @default.
- W3006500171 hasConceptScore W3006500171C81917197 @default.
- W3006500171 hasLocation W30065001711 @default.
- W3006500171 hasOpenAccess W3006500171 @default.
- W3006500171 hasPrimaryLocation W30065001711 @default.
- W3006500171 hasRelatedWork W1970097186 @default.
- W3006500171 hasRelatedWork W2052471706 @default.
- W3006500171 hasRelatedWork W2116383979 @default.
- W3006500171 hasRelatedWork W2531012417 @default.
- W3006500171 hasRelatedWork W2532022121 @default.
- W3006500171 hasRelatedWork W2737900586 @default.
- W3006500171 hasRelatedWork W2791100102 @default.
- W3006500171 hasRelatedWork W2920229473 @default.
- W3006500171 hasRelatedWork W2951862331 @default.
- W3006500171 hasRelatedWork W2964046129 @default.
- W3006500171 hasRelatedWork W3013985224 @default.
- W3006500171 hasRelatedWork W3096366381 @default.
- W3006500171 hasRelatedWork W3100895096 @default.
- W3006500171 hasRelatedWork W3107765940 @default.
- W3006500171 hasRelatedWork W3121822496 @default.
- W3006500171 hasRelatedWork W3131803680 @default.
- W3006500171 hasRelatedWork W3148464484 @default.
- W3006500171 hasRelatedWork W3153039375 @default.
- W3006500171 hasRelatedWork W3155687765 @default.
- W3006500171 hasRelatedWork W3162614895 @default.
- W3006500171 isParatext "false" @default.
- W3006500171 isRetracted "false" @default.