Matches in SemOpenAlex for { <https://semopenalex.org/work/W4200202829> ?p ?o ?g. }
Showing items 1 to 62 of
62
with 100 items per page.
- W4200202829 abstract "With the development of machine learning, the issue of privacy leakage has attracted much attention. Member inference attack is an attack method that threatens the privacy of training datasets. It uses the model’s behavior to infer whether the input user record belongs to the training datasets, and then get the user’s private information according to the purpose of the model. This paper studies the member inference attack under the black box model. We design a defense mechanism to make the learning model and the inference attack model learn from each other, and use the gains from the attack model to fine-tune the last layer’s parameters of the learning model. The fine-tuned learning model can reduce the gains from the membership inference attack with less loss of prediction accuracy. We use different datasets to evaluate the defense mechanism on deep neural networks. The results show that when the training accuracy and test accuracy of the learning model convergence are similar, the learning model only losses about 1% of the prediction accuracy, which the accuracy of the member inference attack drops by a maximum of around 20%." @default.
- W4200202829 created "2021-12-31" @default.
- W4200202829 creator A5013725152 @default.
- W4200202829 creator A5039408870 @default.
- W4200202829 creator A5049950325 @default.
- W4200202829 date "2021-10-01" @default.
- W4200202829 modified "2023-10-06" @default.
- W4200202829 title "A Fine-tuning-based Adversarial Network for Member Privacy Preserving" @default.
- W4200202829 doi "https://doi.org/10.1109/nana53684.2021.00082" @default.
- W4200202829 hasPublicationYear "2021" @default.
- W4200202829 type Work @default.
- W4200202829 citedByCount "0" @default.
- W4200202829 crossrefType "proceedings-article" @default.
- W4200202829 hasAuthorship W4200202829A5013725152 @default.
- W4200202829 hasAuthorship W4200202829A5039408870 @default.
- W4200202829 hasAuthorship W4200202829A5049950325 @default.
- W4200202829 hasConcept C108583219 @default.
- W4200202829 hasConcept C119857082 @default.
- W4200202829 hasConcept C124101348 @default.
- W4200202829 hasConcept C154945302 @default.
- W4200202829 hasConcept C162324750 @default.
- W4200202829 hasConcept C2776214188 @default.
- W4200202829 hasConcept C2777303404 @default.
- W4200202829 hasConcept C2779201187 @default.
- W4200202829 hasConcept C37736160 @default.
- W4200202829 hasConcept C38652104 @default.
- W4200202829 hasConcept C41008148 @default.
- W4200202829 hasConcept C50522688 @default.
- W4200202829 hasConcept C50644808 @default.
- W4200202829 hasConcept C65856478 @default.
- W4200202829 hasConceptScore W4200202829C108583219 @default.
- W4200202829 hasConceptScore W4200202829C119857082 @default.
- W4200202829 hasConceptScore W4200202829C124101348 @default.
- W4200202829 hasConceptScore W4200202829C154945302 @default.
- W4200202829 hasConceptScore W4200202829C162324750 @default.
- W4200202829 hasConceptScore W4200202829C2776214188 @default.
- W4200202829 hasConceptScore W4200202829C2777303404 @default.
- W4200202829 hasConceptScore W4200202829C2779201187 @default.
- W4200202829 hasConceptScore W4200202829C37736160 @default.
- W4200202829 hasConceptScore W4200202829C38652104 @default.
- W4200202829 hasConceptScore W4200202829C41008148 @default.
- W4200202829 hasConceptScore W4200202829C50522688 @default.
- W4200202829 hasConceptScore W4200202829C50644808 @default.
- W4200202829 hasConceptScore W4200202829C65856478 @default.
- W4200202829 hasFunder F4320321001 @default.
- W4200202829 hasFunder F4320335787 @default.
- W4200202829 hasLocation W42002028291 @default.
- W4200202829 hasOpenAccess W4200202829 @default.
- W4200202829 hasPrimaryLocation W42002028291 @default.
- W4200202829 hasRelatedWork W14292303 @default.
- W4200202829 hasRelatedWork W15017807 @default.
- W4200202829 hasRelatedWork W3449668 @default.
- W4200202829 hasRelatedWork W7554620 @default.
- W4200202829 hasRelatedWork W8114543 @default.
- W4200202829 hasRelatedWork W8677904 @default.
- W4200202829 hasRelatedWork W8744369 @default.
- W4200202829 hasRelatedWork W9657784 @default.
- W4200202829 hasRelatedWork W9930909 @default.
- W4200202829 hasRelatedWork W4030176 @default.
- W4200202829 isParatext "false" @default.
- W4200202829 isRetracted "false" @default.
- W4200202829 workType "article" @default.