Matches in SemOpenAlex for { <https://semopenalex.org/work/W3102736503> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W3102736503 endingPage "14422" @default.
- W3102736503 startingPage "14410" @default.
- W3102736503 abstract "Kernel methods provide an elegant and principled approach to nonparametric learning, but so far could hardly be used in large scale problems, since naive implementations scale poorly with data size. Recent advances have shown the benefits of a number of algorithmic ideas, for example combining optimization, numerical linear algebra and random projections. Here, we push these efforts further to develop and test a solver that takes full advantage of GPU hardware. Towards this end, we designed a preconditioned gradient solver for kernel methods exploiting both GPU acceleration and parallelization with multiple GPUs, implementing out-of-core variants of common linear algebra operations to guarantee optimal hardware utilization. Further, we optimize the numerical precision of different operations and maximize efficiency of matrix-vector multiplications. As a result we can experimentally show dramatic speedups on datasets with billions of points, while still guaranteeing state of the art performance. Additionally, we make our software available as an easy to use library." @default.
- W3102736503 created "2020-11-23" @default.
- W3102736503 creator A5013505114 @default.
- W3102736503 creator A5014399020 @default.
- W3102736503 creator A5019263367 @default.
- W3102736503 creator A5061220999 @default.
- W3102736503 date "2020-01-01" @default.
- W3102736503 modified "2023-09-24" @default.
- W3102736503 title "Kernel Methods Through the Roof: Handling Billions of Points Efficiently" @default.
- W3102736503 hasPublicationYear "2020" @default.
- W3102736503 type Work @default.
- W3102736503 sameAs 3102736503 @default.
- W3102736503 citedByCount "13" @default.
- W3102736503 countsByYear W31027365032020 @default.
- W3102736503 countsByYear W31027365032021 @default.
- W3102736503 crossrefType "proceedings-article" @default.
- W3102736503 hasAuthorship W3102736503A5013505114 @default.
- W3102736503 hasAuthorship W3102736503A5014399020 @default.
- W3102736503 hasAuthorship W3102736503A5019263367 @default.
- W3102736503 hasAuthorship W3102736503A5061220999 @default.
- W3102736503 hasConcept C113775141 @default.
- W3102736503 hasConcept C114614502 @default.
- W3102736503 hasConcept C134306372 @default.
- W3102736503 hasConcept C139352143 @default.
- W3102736503 hasConcept C163834973 @default.
- W3102736503 hasConcept C173608175 @default.
- W3102736503 hasConcept C199360897 @default.
- W3102736503 hasConcept C2524010 @default.
- W3102736503 hasConcept C26713055 @default.
- W3102736503 hasConcept C2778119891 @default.
- W3102736503 hasConcept C2778770139 @default.
- W3102736503 hasConcept C33923547 @default.
- W3102736503 hasConcept C41008148 @default.
- W3102736503 hasConcept C459310 @default.
- W3102736503 hasConcept C6802819 @default.
- W3102736503 hasConcept C74193536 @default.
- W3102736503 hasConcept C80444323 @default.
- W3102736503 hasConceptScore W3102736503C113775141 @default.
- W3102736503 hasConceptScore W3102736503C114614502 @default.
- W3102736503 hasConceptScore W3102736503C134306372 @default.
- W3102736503 hasConceptScore W3102736503C139352143 @default.
- W3102736503 hasConceptScore W3102736503C163834973 @default.
- W3102736503 hasConceptScore W3102736503C173608175 @default.
- W3102736503 hasConceptScore W3102736503C199360897 @default.
- W3102736503 hasConceptScore W3102736503C2524010 @default.
- W3102736503 hasConceptScore W3102736503C26713055 @default.
- W3102736503 hasConceptScore W3102736503C2778119891 @default.
- W3102736503 hasConceptScore W3102736503C2778770139 @default.
- W3102736503 hasConceptScore W3102736503C33923547 @default.
- W3102736503 hasConceptScore W3102736503C41008148 @default.
- W3102736503 hasConceptScore W3102736503C459310 @default.
- W3102736503 hasConceptScore W3102736503C6802819 @default.
- W3102736503 hasConceptScore W3102736503C74193536 @default.
- W3102736503 hasConceptScore W3102736503C80444323 @default.
- W3102736503 hasLocation W31027365031 @default.
- W3102736503 hasOpenAccess W3102736503 @default.
- W3102736503 hasPrimaryLocation W31027365031 @default.
- W3102736503 hasRelatedWork W106544777 @default.
- W3102736503 hasRelatedWork W1560724230 @default.
- W3102736503 hasRelatedWork W1788526377 @default.
- W3102736503 hasRelatedWork W1972783048 @default.
- W3102736503 hasRelatedWork W2112545207 @default.
- W3102736503 hasRelatedWork W2160773798 @default.
- W3102736503 hasRelatedWork W2194775991 @default.
- W3102736503 hasRelatedWork W2246456778 @default.
- W3102736503 hasRelatedWork W2599987010 @default.
- W3102736503 hasRelatedWork W2608721507 @default.
- W3102736503 hasRelatedWork W2617819327 @default.
- W3102736503 hasRelatedWork W2963674301 @default.
- W3102736503 hasRelatedWork W2963899745 @default.
- W3102736503 hasRelatedWork W3005868431 @default.
- W3102736503 hasRelatedWork W3036772641 @default.
- W3102736503 hasRelatedWork W3051669108 @default.
- W3102736503 hasRelatedWork W3120934634 @default.
- W3102736503 hasRelatedWork W3176486103 @default.
- W3102736503 hasRelatedWork W3201295431 @default.
- W3102736503 hasRelatedWork W91577598 @default.
- W3102736503 hasVolume "33" @default.
- W3102736503 isParatext "false" @default.
- W3102736503 isRetracted "false" @default.
- W3102736503 magId "3102736503" @default.
- W3102736503 workType "article" @default.