Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387150145> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W4387150145 abstract "To protect users' right to be forgotten in federated learning, federated unlearning aims at eliminating the impact of leaving users' data on the global learned model. The current research in federated unlearning mainly concentrated on developing effective and efficient unlearning techniques. However, the issue of incentivizing valuable users to remain engaged and preventing their data from being unlearned is still under-explored, yet important to the unlearned model performance. This paper focuses on the incentive issue and develops an incentive mechanism for federated learning and unlearning. We first characterize the leaving users' impact on the global model accuracy and the required communication rounds for unlearning. Building on these results, we propose a four-stage game to capture the interaction and information updates during the learning and unlearning process. A key contribution is to summarize users' multi-dimensional private information into one-dimensional metrics to guide the incentive design. We show that users who incur high costs and experience significant training losses are more likely to discontinue their engagement through federated unlearning. The server tends to retain users who make substantial contributions to the model but has a trade-off on users' training losses, as large training losses of retained users increase privacy costs but decrease unlearning costs. The numerical results demonstrate the necessity of unlearning incentives for retaining valuable leaving users, and also show that our proposed mechanisms decrease the server's cost by up to 53.91% compared to state-of-the-art benchmarks." @default.
- W4387150145 created "2023-09-29" @default.
- W4387150145 creator A5000020871 @default.
- W4387150145 creator A5004040268 @default.
- W4387150145 creator A5066234112 @default.
- W4387150145 creator A5085511405 @default.
- W4387150145 date "2023-10-16" @default.
- W4387150145 modified "2023-10-18" @default.
- W4387150145 title "Incentive Mechanism Design for Federated Learning and Unlearning" @default.
- W4387150145 cites W2069119931 @default.
- W4387150145 cites W2744999500 @default.
- W4387150145 cites W2934719948 @default.
- W4387150145 cites W3091870957 @default.
- W4387150145 cites W3112877893 @default.
- W4387150145 cites W3205478485 @default.
- W4387150145 doi "https://doi.org/10.1145/3565287.3610269" @default.
- W4387150145 hasPublicationYear "2023" @default.
- W4387150145 type Work @default.
- W4387150145 citedByCount "0" @default.
- W4387150145 crossrefType "proceedings-article" @default.
- W4387150145 hasAuthorship W4387150145A5000020871 @default.
- W4387150145 hasAuthorship W4387150145A5004040268 @default.
- W4387150145 hasAuthorship W4387150145A5066234112 @default.
- W4387150145 hasAuthorship W4387150145A5085511405 @default.
- W4387150145 hasConcept C111472728 @default.
- W4387150145 hasConcept C111919701 @default.
- W4387150145 hasConcept C11413529 @default.
- W4387150145 hasConcept C138885662 @default.
- W4387150145 hasConcept C154945302 @default.
- W4387150145 hasConcept C162324750 @default.
- W4387150145 hasConcept C175444787 @default.
- W4387150145 hasConcept C26517878 @default.
- W4387150145 hasConcept C29122968 @default.
- W4387150145 hasConcept C2992525071 @default.
- W4387150145 hasConcept C38652104 @default.
- W4387150145 hasConcept C41008148 @default.
- W4387150145 hasConcept C48103436 @default.
- W4387150145 hasConcept C56739046 @default.
- W4387150145 hasConcept C89611455 @default.
- W4387150145 hasConcept C98045186 @default.
- W4387150145 hasConceptScore W4387150145C111472728 @default.
- W4387150145 hasConceptScore W4387150145C111919701 @default.
- W4387150145 hasConceptScore W4387150145C11413529 @default.
- W4387150145 hasConceptScore W4387150145C138885662 @default.
- W4387150145 hasConceptScore W4387150145C154945302 @default.
- W4387150145 hasConceptScore W4387150145C162324750 @default.
- W4387150145 hasConceptScore W4387150145C175444787 @default.
- W4387150145 hasConceptScore W4387150145C26517878 @default.
- W4387150145 hasConceptScore W4387150145C29122968 @default.
- W4387150145 hasConceptScore W4387150145C2992525071 @default.
- W4387150145 hasConceptScore W4387150145C38652104 @default.
- W4387150145 hasConceptScore W4387150145C41008148 @default.
- W4387150145 hasConceptScore W4387150145C48103436 @default.
- W4387150145 hasConceptScore W4387150145C56739046 @default.
- W4387150145 hasConceptScore W4387150145C89611455 @default.
- W4387150145 hasConceptScore W4387150145C98045186 @default.
- W4387150145 hasLocation W43871501451 @default.
- W4387150145 hasOpenAccess W4387150145 @default.
- W4387150145 hasPrimaryLocation W43871501451 @default.
- W4387150145 hasRelatedWork W2037427821 @default.
- W4387150145 hasRelatedWork W2355932053 @default.
- W4387150145 hasRelatedWork W2366296021 @default.
- W4387150145 hasRelatedWork W2367862541 @default.
- W4387150145 hasRelatedWork W2371983804 @default.
- W4387150145 hasRelatedWork W2375027023 @default.
- W4387150145 hasRelatedWork W2382571223 @default.
- W4387150145 hasRelatedWork W2390712896 @default.
- W4387150145 hasRelatedWork W2391671143 @default.
- W4387150145 hasRelatedWork W2392571533 @default.
- W4387150145 isParatext "false" @default.
- W4387150145 isRetracted "false" @default.
- W4387150145 workType "article" @default.