FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3004195290> ?p ?o ?g. }

• W3004195290 endingPage "18" @default.
• W3004195290 startingPage "1" @default.
• W3004195290 abstract "The remarkable success of machine learning has fostered a growing number of cloud-based intelligent services for mobile users. Such a service requires a user to send data, e.g. image, voice and video, to the provider, which presents a serious challenge to user privacy. To address this, prior works either obfuscate the data, e.g. add noise and remove identity information, or send representations extracted from the data, e.g. anonymized features. They struggle to balance between the service utility and data privacy because obfuscated data reduces utility and extracted representation may still reveal sensitive information. This work departs from prior works in methodology: we leverage adversarial learning to a better balance between privacy and utility. We design a textit{representation encoder} that generates the feature representations to optimize against the privacy disclosure risk of sensitive information (a measure of privacy) by the textit{privacy adversaries}, and concurrently optimize with the task inference accuracy (a measure of utility) by the textit{utility discriminator}. The result is the privacy adversarial network (systemname), a novel deep model with the new training algorithm, that can automatically learn representations from the raw data. Intuitively, PAN adversarially forces the extracted representations to only convey the information required by the target task. Surprisingly, this constitutes an implicit regularization that actually improves task accuracy. As a result, PAN achieves better utility and better privacy at the same time! We report extensive experiments on six popular datasets and demonstrate the superiority of systemname compared with alternative methods reported in prior work." @default.
• W3004195290 created "2020-02-07" @default.
• W3004195290 creator A5004062770 @default.
• W3004195290 creator A5016399094 @default.
• W3004195290 creator A5024993683 @default.
• W3004195290 creator A5041610838 @default.
• W3004195290 date "2019-12-11" @default.
• W3004195290 modified "2023-10-04" @default.
• W3004195290 title "Privacy Adversarial Network" @default.
• W3004195290 cites W1915485278 @default.
• W3004195290 cites W2009611335 @default.
• W3004195290 cites W2072566913 @default.
• W3004195290 cites W2092985495 @default.
• W3004195290 cites W2108598243 @default.
• W3004195290 cites W2293078015 @default.
• W3004195290 cites W2408468826 @default.
• W3004195290 cites W2473418344 @default.
• W3004195290 cites W2520881573 @default.
• W3004195290 cites W2536502423 @default.
• W3004195290 cites W2605289356 @default.
• W3004195290 cites W2626533970 @default.
• W3004195290 cites W2734500670 @default.
• W3004195290 cites W2860338957 @default.
• W3004195290 cites W2884985453 @default.
• W3004195290 cites W2906993533 @default.
• W3004195290 cites W2962914239 @default.
• W3004195290 cites W2963278610 @default.
• W3004195290 cites W3101314853 @default.
• W3004195290 cites W4205228770 @default.
• W3004195290 doi "https://doi.org/10.1145/3369816" @default.
• W3004195290 hasPublicationYear "2019" @default.
• W3004195290 type Work @default.
• W3004195290 sameAs 3004195290 @default.
• W3004195290 citedByCount "15" @default.
• W3004195290 countsByYear W30041952902019 @default.
• W3004195290 countsByYear W30041952902020 @default.
• W3004195290 countsByYear W30041952902021 @default.
• W3004195290 countsByYear W30041952902022 @default.
• W3004195290 countsByYear W30041952902023 @default.
• W3004195290 crossrefType "journal-article" @default.
• W3004195290 hasAuthorship W3004195290A5004062770 @default.
• W3004195290 hasAuthorship W3004195290A5016399094 @default.
• W3004195290 hasAuthorship W3004195290A5024993683 @default.
• W3004195290 hasAuthorship W3004195290A5041610838 @default.
• W3004195290 hasBestOaLocation W30041952902 @default.
• W3004195290 hasConcept C119857082 @default.
• W3004195290 hasConcept C123201435 @default.
• W3004195290 hasConcept C124101348 @default.
• W3004195290 hasConcept C153083717 @default.
• W3004195290 hasConcept C154945302 @default.
• W3004195290 hasConcept C162324750 @default.
• W3004195290 hasConcept C17744445 @default.
• W3004195290 hasConcept C187736073 @default.
• W3004195290 hasConcept C199539241 @default.
• W3004195290 hasConcept C23123220 @default.
• W3004195290 hasConcept C26517878 @default.
• W3004195290 hasConcept C2776214188 @default.
• W3004195290 hasConcept C2776359362 @default.
• W3004195290 hasConcept C2779803651 @default.
• W3004195290 hasConcept C2780451532 @default.
• W3004195290 hasConcept C37736160 @default.
• W3004195290 hasConcept C38652104 @default.
• W3004195290 hasConcept C41008148 @default.
• W3004195290 hasConcept C59404180 @default.
• W3004195290 hasConcept C76155785 @default.
• W3004195290 hasConcept C94625758 @default.
• W3004195290 hasConcept C94915269 @default.
• W3004195290 hasConceptScore W3004195290C119857082 @default.
• W3004195290 hasConceptScore W3004195290C123201435 @default.
• W3004195290 hasConceptScore W3004195290C124101348 @default.
• W3004195290 hasConceptScore W3004195290C153083717 @default.
• W3004195290 hasConceptScore W3004195290C154945302 @default.
• W3004195290 hasConceptScore W3004195290C162324750 @default.
• W3004195290 hasConceptScore W3004195290C17744445 @default.
• W3004195290 hasConceptScore W3004195290C187736073 @default.
• W3004195290 hasConceptScore W3004195290C199539241 @default.
• W3004195290 hasConceptScore W3004195290C23123220 @default.
• W3004195290 hasConceptScore W3004195290C26517878 @default.
• W3004195290 hasConceptScore W3004195290C2776214188 @default.
• W3004195290 hasConceptScore W3004195290C2776359362 @default.
• W3004195290 hasConceptScore W3004195290C2779803651 @default.
• W3004195290 hasConceptScore W3004195290C2780451532 @default.
• W3004195290 hasConceptScore W3004195290C37736160 @default.
• W3004195290 hasConceptScore W3004195290C38652104 @default.
• W3004195290 hasConceptScore W3004195290C41008148 @default.
• W3004195290 hasConceptScore W3004195290C59404180 @default.
• W3004195290 hasConceptScore W3004195290C76155785 @default.
• W3004195290 hasConceptScore W3004195290C94625758 @default.
• W3004195290 hasConceptScore W3004195290C94915269 @default.
• W3004195290 hasFunder F4320306076 @default.
• W3004195290 hasFunder F4320321001 @default.
• W3004195290 hasFunder F4320335777 @default.
• W3004195290 hasIssue "4" @default.
• W3004195290 hasLocation W30041952901 @default.
• W3004195290 hasLocation W30041952902 @default.
• W3004195290 hasOpenAccess W3004195290 @default.
• W3004195290 hasPrimaryLocation W30041952901 @default.
• W3004195290 hasRelatedWork W2891261571 @default.

• next