Matches in SemOpenAlex for { <https://semopenalex.org/work/W2989336765> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W2989336765 abstract "Online kernel selection is a more complex problem compared with offline kernel selection, which intermixes training and selection at each round and requires a sublinear regret and low computational complexities. But existing online kernel selection approaches have at least linear time and space complexities at each round with respect to the number of rounds, or lack sublinear regret guarantees for an uncountably infinite number of candidate kernels. To address these issues, we propose a novel online kernel selection approach using tensor sketching, which has constant computational complexities at each round and enjoys a sublinear regret bound for an uncountably infinite number of candidate kernels. We represent the data using the tensor products and construct data sketches using the Taylor series and the Count Sketch matrices, which yields a sketched reproducing kernel Hilbert space (SRKHS). Then we update the optimal kernels and the hypotheses using online gradient descent in SRKHS. We prove that the kernel corresponding to SRKHS satisfies the reproducing property, the hypotheses in SRKHS are convex with respect to the kernel parameter, and the proposed online kernel selection approach in SRKHS enjoys a regret bound of order $O(sqrtT )$ for an uncountably infinite number of candidate kernels, which is optimal for a convex loss function, where T is the number of rounds. By the fast Fourier transform, the hypotheses in SRKHS can be computed in a quasilinear time complexity and a logarithmic space complexity with respect to the sketch size at each round, where the sketch size is a constant. Experimental results demonstrate that our online kernel selection approach is more accurate and efficient for online kernel learning on high dimension data." @default.
- W2989336765 created "2019-11-22" @default.
- W2989336765 creator A5002318539 @default.
- W2989336765 creator A5089843351 @default.
- W2989336765 date "2019-11-03" @default.
- W2989336765 modified "2023-09-22" @default.
- W2989336765 title "Online Kernel Selection via Tensor Sketching" @default.
- W2989336765 cites W1492499319 @default.
- W2989336765 cites W1493892051 @default.
- W2989336765 cites W1510073064 @default.
- W2989336765 cites W1570963478 @default.
- W2989336765 cites W1677330902 @default.
- W2989336765 cites W1993288992 @default.
- W2989336765 cites W2064379477 @default.
- W2989336765 cites W2077723394 @default.
- W2989336765 cites W2146897752 @default.
- W2989336765 cites W2153918971 @default.
- W2989336765 cites W2201600774 @default.
- W2989336765 cites W2408432900 @default.
- W2989336765 cites W2513180554 @default.
- W2989336765 cites W2808275179 @default.
- W2989336765 cites W2903088943 @default.
- W2989336765 doi "https://doi.org/10.1145/3357384.3358019" @default.
- W2989336765 hasPublicationYear "2019" @default.
- W2989336765 type Work @default.
- W2989336765 sameAs 2989336765 @default.
- W2989336765 citedByCount "0" @default.
- W2989336765 crossrefType "proceedings-article" @default.
- W2989336765 hasAuthorship W2989336765A5002318539 @default.
- W2989336765 hasAuthorship W2989336765A5089843351 @default.
- W2989336765 hasConcept C154945302 @default.
- W2989336765 hasConcept C155281189 @default.
- W2989336765 hasConcept C202444582 @default.
- W2989336765 hasConcept C33923547 @default.
- W2989336765 hasConcept C41008148 @default.
- W2989336765 hasConcept C74193536 @default.
- W2989336765 hasConcept C81917197 @default.
- W2989336765 hasConceptScore W2989336765C154945302 @default.
- W2989336765 hasConceptScore W2989336765C155281189 @default.
- W2989336765 hasConceptScore W2989336765C202444582 @default.
- W2989336765 hasConceptScore W2989336765C33923547 @default.
- W2989336765 hasConceptScore W2989336765C41008148 @default.
- W2989336765 hasConceptScore W2989336765C74193536 @default.
- W2989336765 hasConceptScore W2989336765C81917197 @default.
- W2989336765 hasLocation W29893367651 @default.
- W2989336765 hasOpenAccess W2989336765 @default.
- W2989336765 hasPrimaryLocation W29893367651 @default.
- W2989336765 hasRelatedWork W1979779627 @default.
- W2989336765 hasRelatedWork W1990422213 @default.
- W2989336765 hasRelatedWork W2085949256 @default.
- W2989336765 hasRelatedWork W2115461783 @default.
- W2989336765 hasRelatedWork W2606073172 @default.
- W2989336765 hasRelatedWork W2970622411 @default.
- W2989336765 hasRelatedWork W3107474891 @default.
- W2989336765 hasRelatedWork W4230772827 @default.
- W2989336765 hasRelatedWork W4250047567 @default.
- W2989336765 hasRelatedWork W2751504412 @default.
- W2989336765 isParatext "false" @default.
- W2989336765 isRetracted "false" @default.
- W2989336765 magId "2989336765" @default.
- W2989336765 workType "article" @default.