FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912012050> ?p ?o ?g. }

• W2912012050 abstract "The remarkable success of machine learning, especially deep learning, has produced a variety of cloud-based services for mobile users. Such services require an end user to send data to the service provider, which presents a serious challenge to end-user privacy. To address this concern, prior works either add noise to the data or send features extracted from the raw data. They struggle to balance between the utility and privacy because added noise reduces utility and raw data can be reconstructed from extracted features. This work represents a methodical departure from prior works: we balance between a measure of privacy and another of utility by leveraging adversarial learning to find a sweeter tradeoff. We design an encoder that optimizes against the reconstruction error (a measure of privacy), adversarially by a Decoder, and the inference accuracy (a measure of utility) by a Classifier. The result is RAN, a novel deep model with a new training algorithm that automatically extracts features for classification that are both private and useful. It turns out that adversarially forcing the extracted features to only conveys the intended information required by classification leads to an implicit regularization leading to better classification accuracy than the original model which completely ignores privacy. Thus, we achieve better privacy with better utility, a surprising possibility in machine learning! We conducted extensive experiments on five popular datasets over four training schemes, and demonstrate the superiority of RAN compared with existing alternatives." @default.
• W2912012050 created "2019-02-21" @default.
• W2912012050 creator A5004062770 @default.
• W2912012050 creator A5016399094 @default.
• W2912012050 creator A5024993683 @default.
• W2912012050 creator A5041610838 @default.
• W2912012050 date "2018-09-27" @default.
• W2912012050 modified "2023-09-25" @default.
• W2912012050 title "Better Accuracy with Quantified Privacy: Representations Learned via Reconstructive Adversarial Network" @default.
• W2912012050 cites W1522301498 @default.
• W2912012050 cites W1915485278 @default.
• W2912012050 cites W2009611335 @default.
• W2912012050 cites W2027595342 @default.
• W2912012050 cites W2072566913 @default.
• W2912012050 cites W2099471712 @default.
• W2912012050 cites W2108598243 @default.
• W2912012050 cites W2173520492 @default.
• W2912012050 cites W2293078015 @default.
• W2912012050 cites W2317339301 @default.
• W2912012050 cites W2408468826 @default.
• W2912012050 cites W2467604901 @default.
• W2912012050 cites W2473418344 @default.
• W2912012050 cites W2520881573 @default.
• W2912012050 cites W2536502423 @default.
• W2912012050 cites W2558531213 @default.
• W2912012050 cites W2597998853 @default.
• W2912012050 cites W2626533970 @default.
• W2912012050 cites W2727248480 @default.
• W2912012050 cites W2734500670 @default.
• W2912012050 cites W2949117887 @default.
• W2912012050 cites W2962914239 @default.
• W2912012050 cites W31376329 @default.
• W2912012050 hasPublicationYear "2018" @default.
• W2912012050 type Work @default.
• W2912012050 sameAs 2912012050 @default.
• W2912012050 citedByCount "4" @default.
• W2912012050 countsByYear W29120120502019 @default.
• W2912012050 countsByYear W29120120502020 @default.
• W2912012050 countsByYear W29120120502021 @default.
• W2912012050 crossrefType "posted-content" @default.
• W2912012050 hasAuthorship W2912012050A5004062770 @default.
• W2912012050 hasAuthorship W2912012050A5016399094 @default.
• W2912012050 hasAuthorship W2912012050A5024993683 @default.
• W2912012050 hasAuthorship W2912012050A5041610838 @default.
• W2912012050 hasConcept C108583219 @default.
• W2912012050 hasConcept C111919701 @default.
• W2912012050 hasConcept C119857082 @default.
• W2912012050 hasConcept C123201435 @default.
• W2912012050 hasConcept C124101348 @default.
• W2912012050 hasConcept C132964779 @default.
• W2912012050 hasConcept C154945302 @default.
• W2912012050 hasConcept C199360897 @default.
• W2912012050 hasConcept C23130292 @default.
• W2912012050 hasConcept C2776135515 @default.
• W2912012050 hasConcept C2776214188 @default.
• W2912012050 hasConcept C37736160 @default.
• W2912012050 hasConcept C38652104 @default.
• W2912012050 hasConcept C41008148 @default.
• W2912012050 hasConcept C79974875 @default.
• W2912012050 hasConcept C95623464 @default.
• W2912012050 hasConceptScore W2912012050C108583219 @default.
• W2912012050 hasConceptScore W2912012050C111919701 @default.
• W2912012050 hasConceptScore W2912012050C119857082 @default.
• W2912012050 hasConceptScore W2912012050C123201435 @default.
• W2912012050 hasConceptScore W2912012050C124101348 @default.
• W2912012050 hasConceptScore W2912012050C132964779 @default.
• W2912012050 hasConceptScore W2912012050C154945302 @default.
• W2912012050 hasConceptScore W2912012050C199360897 @default.
• W2912012050 hasConceptScore W2912012050C23130292 @default.
• W2912012050 hasConceptScore W2912012050C2776135515 @default.
• W2912012050 hasConceptScore W2912012050C2776214188 @default.
• W2912012050 hasConceptScore W2912012050C37736160 @default.
• W2912012050 hasConceptScore W2912012050C38652104 @default.
• W2912012050 hasConceptScore W2912012050C41008148 @default.
• W2912012050 hasConceptScore W2912012050C79974875 @default.
• W2912012050 hasConceptScore W2912012050C95623464 @default.
• W2912012050 hasLocation W29120120501 @default.
• W2912012050 hasOpenAccess W2912012050 @default.
• W2912012050 hasPrimaryLocation W29120120501 @default.
• W2912012050 hasRelatedWork W1585030920 @default.
• W2912012050 hasRelatedWork W2766255512 @default.
• W2912012050 hasRelatedWork W2900598269 @default.
• W2912012050 hasRelatedWork W2904885070 @default.
• W2912012050 hasRelatedWork W2952650103 @default.
• W2912012050 hasRelatedWork W2972933328 @default.
• W2912012050 hasRelatedWork W2991272391 @default.
• W2912012050 hasRelatedWork W2996288166 @default.
• W2912012050 hasRelatedWork W3003784440 @default.
• W2912012050 hasRelatedWork W3005002459 @default.
• W2912012050 hasRelatedWork W3007228146 @default.
• W2912012050 hasRelatedWork W3035174002 @default.
• W2912012050 hasRelatedWork W3046565236 @default.
• W2912012050 hasRelatedWork W3048511172 @default.
• W2912012050 hasRelatedWork W3104406626 @default.
• W2912012050 hasRelatedWork W3119876694 @default.
• W2912012050 hasRelatedWork W3119922393 @default.
• W2912012050 hasRelatedWork W3138774967 @default.
• W2912012050 hasRelatedWork W3161996757 @default.
• W2912012050 hasRelatedWork W3015983782 @default.
• W2912012050 isParatext "false" @default.

• next