FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387490675> ?p ?o ?g. }

• W4387490675 endingPage "111225" @default.
• W4387490675 startingPage "111212" @default.
• W4387490675 abstract "Federated learning (FL), which not only protects data security and privacy, but also training models on distributed devices, receives much attention in the literature. Traditionally, stochastic gradient descent (SGD) is used in FL since its excellent empirical performance, but user’s private information can still be leaked by analyzing weight updates during FL iterations. Differential privacy (DP) is an effective way to solve this privacy leakage problem, which adds noise to the user’s data gradient and this artificial noise helps to prevent information leakage. However, note that the SGD based FL with DP is not yet investigated with a comprehensive theoretical analysis considering privacy and data utility jointly, especially from the information-theoretic aspect. In this paper, we investigate the FL in the presence of mutual information based DP (MI-DP). Specifically, first, Gaussian DP mechanism is applied to either clients or central server of the FL model, and privacy and utility of the FL model are characterized by conditional mutual information and distortion, respectively. For a given privacy budget, we establish lower bounds on the variance of the Gaussian noise added to clients or central server of the FL model, and show that the utility of the global model remains the same for both cases. Next, we study the privacy-utility trade-off problem by considering a more general case, where both the model parameter and the privacy requirement of the clients are flexible. A privacy-preserving scheme is proposed, which maximizes the utility of the global model while different privacy requirements of all clients are preserved. Finally, the results of this paper are further explained by experimental results." @default.
• W4387490675 created "2023-10-11" @default.
• W4387490675 creator A5037259398 @default.
• W4387490675 creator A5055423191 @default.
• W4387490675 date "2023-01-01" @default.
• W4387490675 modified "2023-10-18" @default.
• W4387490675 title "Minimum Gaussian Noise Variance of Federated Learning in the Presence of Mutual Information Based Differential Privacy" @default.
• W4387490675 cites W1649611255 @default.
• W4387490675 cites W2077145963 @default.
• W4387490675 cites W2112796928 @default.
• W4387490675 cites W2113792377 @default.
• W4387490675 cites W2130964760 @default.
• W4387490675 cites W2515654213 @default.
• W4387490675 cites W2535690855 @default.
• W4387490675 cites W2564029303 @default.
• W4387490675 cites W2591882872 @default.
• W4387490675 cites W2911978475 @default.
• W4387490675 cites W2912213068 @default.
• W4387490675 cites W2963456518 @default.
• W4387490675 cites W3027749727 @default.
• W4387490675 cites W3099980085 @default.
• W4387490675 cites W3180608480 @default.
• W4387490675 cites W4206320562 @default.
• W4387490675 cites W4300840296 @default.
• W4387490675 cites W4379618860 @default.
• W4387490675 doi "https://doi.org/10.1109/access.2023.3323020" @default.
• W4387490675 hasPublicationYear "2023" @default.
• W4387490675 type Work @default.
• W4387490675 citedByCount "0" @default.
• W4387490675 crossrefType "journal-article" @default.
• W4387490675 hasAuthorship W4387490675A5037259398 @default.
• W4387490675 hasAuthorship W4387490675A5055423191 @default.
• W4387490675 hasBestOaLocation W43874906751 @default.
• W4387490675 hasConcept C115961682 @default.
• W4387490675 hasConcept C121332964 @default.
• W4387490675 hasConcept C121955636 @default.
• W4387490675 hasConcept C123201435 @default.
• W4387490675 hasConcept C124101348 @default.
• W4387490675 hasConcept C127162648 @default.
• W4387490675 hasConcept C137822555 @default.
• W4387490675 hasConcept C144133560 @default.
• W4387490675 hasConcept C152139883 @default.
• W4387490675 hasConcept C154945302 @default.
• W4387490675 hasConcept C163716315 @default.
• W4387490675 hasConcept C196083921 @default.
• W4387490675 hasConcept C206688291 @default.
• W4387490675 hasConcept C23130292 @default.
• W4387490675 hasConcept C2779201187 @default.
• W4387490675 hasConcept C2780909371 @default.
• W4387490675 hasConcept C31258907 @default.
• W4387490675 hasConcept C38652104 @default.
• W4387490675 hasConcept C41008148 @default.
• W4387490675 hasConcept C4199805 @default.
• W4387490675 hasConcept C47798520 @default.
• W4387490675 hasConcept C50644808 @default.
• W4387490675 hasConcept C62520636 @default.
• W4387490675 hasConcept C99221444 @default.
• W4387490675 hasConcept C99498987 @default.
• W4387490675 hasConceptScore W4387490675C115961682 @default.
• W4387490675 hasConceptScore W4387490675C121332964 @default.
• W4387490675 hasConceptScore W4387490675C121955636 @default.
• W4387490675 hasConceptScore W4387490675C123201435 @default.
• W4387490675 hasConceptScore W4387490675C124101348 @default.
• W4387490675 hasConceptScore W4387490675C127162648 @default.
• W4387490675 hasConceptScore W4387490675C137822555 @default.
• W4387490675 hasConceptScore W4387490675C144133560 @default.
• W4387490675 hasConceptScore W4387490675C152139883 @default.
• W4387490675 hasConceptScore W4387490675C154945302 @default.
• W4387490675 hasConceptScore W4387490675C163716315 @default.
• W4387490675 hasConceptScore W4387490675C196083921 @default.
• W4387490675 hasConceptScore W4387490675C206688291 @default.
• W4387490675 hasConceptScore W4387490675C23130292 @default.
• W4387490675 hasConceptScore W4387490675C2779201187 @default.
• W4387490675 hasConceptScore W4387490675C2780909371 @default.
• W4387490675 hasConceptScore W4387490675C31258907 @default.
• W4387490675 hasConceptScore W4387490675C38652104 @default.
• W4387490675 hasConceptScore W4387490675C41008148 @default.
• W4387490675 hasConceptScore W4387490675C4199805 @default.
• W4387490675 hasConceptScore W4387490675C47798520 @default.
• W4387490675 hasConceptScore W4387490675C50644808 @default.
• W4387490675 hasConceptScore W4387490675C62520636 @default.
• W4387490675 hasConceptScore W4387490675C99221444 @default.
• W4387490675 hasConceptScore W4387490675C99498987 @default.
• W4387490675 hasFunder F4320333335 @default.
• W4387490675 hasLocation W43874906751 @default.
• W4387490675 hasOpenAccess W4387490675 @default.
• W4387490675 hasPrimaryLocation W43874906751 @default.
• W4387490675 hasRelatedWork W2019704260 @default.
• W4387490675 hasRelatedWork W3010824232 @default.
• W4387490675 hasRelatedWork W3093310219 @default.
• W4387490675 hasRelatedWork W3101646702 @default.
• W4387490675 hasRelatedWork W4212899026 @default.
• W4387490675 hasRelatedWork W4296973715 @default.
• W4387490675 hasRelatedWork W4304208041 @default.
• W4387490675 hasRelatedWork W4308633695 @default.
• W4387490675 hasRelatedWork W4361791424 @default.
• W4387490675 hasRelatedWork W4387193529 @default.
• W4387490675 hasVolume "11" @default.

• next