FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3129161255> ?p ?o ?g. }

• W3129161255 abstract "Federated Learning (FL) is an emerging paradigm through which decentralized devices can collaboratively train a common model. However, a serious concern is the leakage of privacy from exchanged gradient information between clients and the parameter server (PS) in FL. To protect gradient information, clients can adopt differential privacy (DP) to add additional noises and distort original gradients before they are uploaded to the PS. Nevertheless, the model accuracy will be significantly impaired by DP noises, making DP impracticable in real systems. In this work, we propose a novel Noise Information Secretly Sharing (NISS) algorithm to alleviate the disturbance of DP noises by sharing negated noises among clients. We theoretically prove that: 1) If clients are trustworthy, DP noises can be perfectly offset on the PS; 2) Clients can easily distort negated DP noises to protect themselves in case that other clients are not totally trustworthy, though the cost lowers model accuracy. NISS is particularly applicable for FL across multiple IoT (Internet of Things) systems, in which all IoT devices need to collaboratively train a model. To verify the effectiveness and the superiority of the NISS algorithm, we conduct experiments with the MNIST and CIFAR-10 datasets. The experiment results verify our analysis and demonstrate that NISS can improve model accuracy by 21% on average and obtain better privacy protection if clients are trustworthy." @default.
• W3129161255 created "2021-02-15" @default.
• W3129161255 creator A5001396358 @default.
• W3129161255 creator A5035113789 @default.
• W3129161255 creator A5041588523 @default.
• W3129161255 creator A5047475186 @default.
• W3129161255 creator A5054798505 @default.
• W3129161255 creator A5081182489 @default.
• W3129161255 creator A5090366405 @default.
• W3129161255 date "2021-01-31" @default.
• W3129161255 modified "2023-10-16" @default.
• W3129161255 title "Gain without Pain: Offsetting DP-injected Nosies Stealthily in Cross-device Federated Learning" @default.
• W3129161255 cites W1510952750 @default.
• W3129161255 cites W2027595342 @default.
• W3129161255 cites W2051434435 @default.
• W3129161255 cites W2056363353 @default.
• W3129161255 cites W2093367651 @default.
• W3129161255 cites W2105103777 @default.
• W3129161255 cites W2111619626 @default.
• W3129161255 cites W2128906841 @default.
• W3129161255 cites W2473418344 @default.
• W3129161255 cites W2530417694 @default.
• W3129161255 cites W2535838896 @default.
• W3129161255 cites W2541884796 @default.
• W3129161255 cites W2591882872 @default.
• W3129161255 cites W2767079719 @default.
• W3129161255 cites W2773194476 @default.
• W3129161255 cites W2777914285 @default.
• W3129161255 cites W2781091734 @default.
• W3129161255 cites W2807006176 @default.
• W3129161255 cites W2836883261 @default.
• W3129161255 cites W2899771611 @default.
• W3129161255 cites W2904308246 @default.
• W3129161255 cites W2912213068 @default.
• W3129161255 cites W2955213239 @default.
• W3129161255 cites W2962883549 @default.
• W3129161255 cites W2963456518 @default.
• W3129161255 cites W2963540401 @default.
• W3129161255 cites W2970408908 @default.
• W3129161255 cites W2995022099 @default.
• W3129161255 cites W3000479830 @default.
• W3129161255 cites W3015535799 @default.
• W3129161255 cites W3016632787 @default.
• W3129161255 cites W3016931901 @default.
• W3129161255 cites W3051272477 @default.
• W3129161255 cites W3084128899 @default.
• W3129161255 cites W3084130864 @default.
• W3129161255 cites W3103802018 @default.
• W3129161255 doi "https://doi.org/10.48550/arxiv.2102.00404" @default.
• W3129161255 hasPublicationYear "2021" @default.
• W3129161255 type Work @default.
• W3129161255 sameAs 3129161255 @default.
• W3129161255 citedByCount "0" @default.
• W3129161255 crossrefType "posted-content" @default.
• W3129161255 hasAuthorship W3129161255A5001396358 @default.
• W3129161255 hasAuthorship W3129161255A5035113789 @default.
• W3129161255 hasAuthorship W3129161255A5041588523 @default.
• W3129161255 hasAuthorship W3129161255A5047475186 @default.
• W3129161255 hasAuthorship W3129161255A5054798505 @default.
• W3129161255 hasAuthorship W3129161255A5081182489 @default.
• W3129161255 hasAuthorship W3129161255A5090366405 @default.
• W3129161255 hasBestOaLocation W31291612551 @default.
• W3129161255 hasConcept C108583219 @default.
• W3129161255 hasConcept C124101348 @default.
• W3129161255 hasConcept C136764020 @default.
• W3129161255 hasConcept C153701036 @default.
• W3129161255 hasConcept C154945302 @default.
• W3129161255 hasConcept C190502265 @default.
• W3129161255 hasConcept C23130292 @default.
• W3129161255 hasConcept C2779201187 @default.
• W3129161255 hasConcept C38652104 @default.
• W3129161255 hasConcept C41008148 @default.
• W3129161255 hasConcept C71901391 @default.
• W3129161255 hasConceptScore W3129161255C108583219 @default.
• W3129161255 hasConceptScore W3129161255C124101348 @default.
• W3129161255 hasConceptScore W3129161255C136764020 @default.
• W3129161255 hasConceptScore W3129161255C153701036 @default.
• W3129161255 hasConceptScore W3129161255C154945302 @default.
• W3129161255 hasConceptScore W3129161255C190502265 @default.
• W3129161255 hasConceptScore W3129161255C23130292 @default.
• W3129161255 hasConceptScore W3129161255C2779201187 @default.
• W3129161255 hasConceptScore W3129161255C38652104 @default.
• W3129161255 hasConceptScore W3129161255C41008148 @default.
• W3129161255 hasConceptScore W3129161255C71901391 @default.
• W3129161255 hasLocation W31291612551 @default.
• W3129161255 hasOpenAccess W3129161255 @default.
• W3129161255 hasPrimaryLocation W31291612551 @default.
• W3129161255 hasRelatedWork W2124572775 @default.
• W3129161255 hasRelatedWork W2285788670 @default.
• W3129161255 hasRelatedWork W2355664989 @default.
• W3129161255 hasRelatedWork W2365093105 @default.
• W3129161255 hasRelatedWork W2371012949 @default.
• W3129161255 hasRelatedWork W2388056151 @default.
• W3129161255 hasRelatedWork W2388811618 @default.
• W3129161255 hasRelatedWork W2947175736 @default.
• W3129161255 hasRelatedWork W3129161255 @default.
• W3129161255 hasRelatedWork W4229783568 @default.
• W3129161255 isParatext "false" @default.
• W3129161255 isRetracted "false" @default.
• W3129161255 magId "3129161255" @default.

• next