FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3025816195> ?p ?o ?g. }

• W3025816195 endingPage "305" @default.
• W3025816195 startingPage "251" @default.
• W3025816195 abstract "Abstract This work addresses the problem of learning from large collections of data with privacy guarantees. The compressive learning framework proposes to deal with the large scale of datasets by compressing them into a single vector of generalized random moments, called a sketch vector, from which the learning task is then performed. We provide sharp bounds on the so-called sensitivity of this sketching mechanism. This allows us to leverage standard techniques to ensure differential privacy—a well-established formalism for defining and quantifying the privacy of a random mechanism—by adding Laplace of Gaussian noise to the sketch. We combine these standard mechanisms with a new feature subsampling mechanism, which reduces the computational cost without damaging privacy. The overall framework is applied to the tasks of Gaussian modeling, k-means clustering and principal component analysis, for which sharp privacy bounds are derived. Empirically, the quality (for subsequent learning) of the compressed representation produced by our mechanism is strongly related with the induced noise level, for which we give analytical expressions." @default.
• W3025816195 created "2020-05-21" @default.
• W3025816195 creator A5028331312 @default.
• W3025816195 creator A5041230403 @default.
• W3025816195 creator A5053892288 @default.
• W3025816195 creator A5073216192 @default.
• W3025816195 creator A5078253058 @default.
• W3025816195 creator A5087376727 @default.
• W3025816195 date "2021-05-15" @default.
• W3025816195 modified "2023-10-03" @default.
• W3025816195 title "Compressive learning with privacy guarantees" @default.
• W3025816195 cites W143004564 @default.
• W3025816195 cites W1602085912 @default.
• W3025816195 cites W1873763122 @default.
• W3025816195 cites W1992926795 @default.
• W3025816195 cites W2010523825 @default.
• W3025816195 cites W2013823004 @default.
• W3025816195 cites W2022858489 @default.
• W3025816195 cites W2024021429 @default.
• W3025816195 cites W2042469398 @default.
• W3025816195 cites W2096870293 @default.
• W3025816195 cites W2103142290 @default.
• W3025816195 cites W2109426455 @default.
• W3025816195 cites W2110868467 @default.
• W3025816195 cites W2115240023 @default.
• W3025816195 cites W2140089781 @default.
• W3025816195 cites W2143075842 @default.
• W3025816195 cites W2151320232 @default.
• W3025816195 cites W2159024459 @default.
• W3025816195 cites W2293934387 @default.
• W3025816195 cites W2402512736 @default.
• W3025816195 cites W2564029303 @default.
• W3025816195 cites W2769610134 @default.
• W3025816195 cites W2794589353 @default.
• W3025816195 cites W2797637497 @default.
• W3025816195 cites W2798310356 @default.
• W3025816195 cites W2889075913 @default.
• W3025816195 cites W2913538792 @default.
• W3025816195 cites W2929827560 @default.
• W3025816195 cites W2963809394 @default.
• W3025816195 cites W3102407811 @default.
• W3025816195 doi "https://doi.org/10.1093/imaiai/iaab005" @default.
• W3025816195 hasPublicationYear "2021" @default.
• W3025816195 type Work @default.
• W3025816195 sameAs 3025816195 @default.
• W3025816195 citedByCount "3" @default.
• W3025816195 countsByYear W30258161952021 @default.
• W3025816195 countsByYear W30258161952022 @default.
• W3025816195 countsByYear W30258161952023 @default.
• W3025816195 crossrefType "journal-article" @default.
• W3025816195 hasAuthorship W3025816195A5028331312 @default.
• W3025816195 hasAuthorship W3025816195A5041230403 @default.
• W3025816195 hasAuthorship W3025816195A5053892288 @default.
• W3025816195 hasAuthorship W3025816195A5073216192 @default.
• W3025816195 hasAuthorship W3025816195A5078253058 @default.
• W3025816195 hasAuthorship W3025816195A5087376727 @default.
• W3025816195 hasBestOaLocation W30258161953 @default.
• W3025816195 hasConcept C11413529 @default.
• W3025816195 hasConcept C119857082 @default.
• W3025816195 hasConcept C121332964 @default.
• W3025816195 hasConcept C124101348 @default.
• W3025816195 hasConcept C153083717 @default.
• W3025816195 hasConcept C154945302 @default.
• W3025816195 hasConcept C163716315 @default.
• W3025816195 hasConcept C23130292 @default.
• W3025816195 hasConcept C27438332 @default.
• W3025816195 hasConcept C2779231336 @default.
• W3025816195 hasConcept C41008148 @default.
• W3025816195 hasConcept C59404180 @default.
• W3025816195 hasConcept C62520636 @default.
• W3025816195 hasConcept C73555534 @default.
• W3025816195 hasConcept C80444323 @default.
• W3025816195 hasConcept C83665646 @default.
• W3025816195 hasConceptScore W3025816195C11413529 @default.
• W3025816195 hasConceptScore W3025816195C119857082 @default.
• W3025816195 hasConceptScore W3025816195C121332964 @default.
• W3025816195 hasConceptScore W3025816195C124101348 @default.
• W3025816195 hasConceptScore W3025816195C153083717 @default.
• W3025816195 hasConceptScore W3025816195C154945302 @default.
• W3025816195 hasConceptScore W3025816195C163716315 @default.
• W3025816195 hasConceptScore W3025816195C23130292 @default.
• W3025816195 hasConceptScore W3025816195C27438332 @default.
• W3025816195 hasConceptScore W3025816195C2779231336 @default.
• W3025816195 hasConceptScore W3025816195C41008148 @default.
• W3025816195 hasConceptScore W3025816195C59404180 @default.
• W3025816195 hasConceptScore W3025816195C62520636 @default.
• W3025816195 hasConceptScore W3025816195C73555534 @default.
• W3025816195 hasConceptScore W3025816195C80444323 @default.
• W3025816195 hasConceptScore W3025816195C83665646 @default.
• W3025816195 hasIssue "1" @default.
• W3025816195 hasLocation W30258161951 @default.
• W3025816195 hasLocation W30258161952 @default.
• W3025816195 hasLocation W30258161953 @default.
• W3025816195 hasLocation W30258161954 @default.
• W3025816195 hasLocation W30258161955 @default.
• W3025816195 hasLocation W30258161956 @default.
• W3025816195 hasLocation W30258161957 @default.
• W3025816195 hasOpenAccess W3025816195 @default.

• next