Matches in SemOpenAlex for { <https://semopenalex.org/work/W2969446490> ?p ?o ?g. }
- W2969446490 abstract "In this paper, we aim to understand the generalization properties of generative adversarial networks (GANs) from a new perspective of privacy protection. Theoretically, we prove that a differentially private learning algorithm used for training the GAN does not overfit to a certain degree, i.e., the generalization gap can be bounded. Moreover, some recent works, such as the Bayesian GAN, can be re-interpreted based on our theoretical insight from privacy protection. Quantitatively, to evaluate the information leakage of well-trained GAN models, we perform various membership attacks on these models. The results show that previous Lipschitz regularization techniques are effective in not only reducing the generalization gap but also alleviating the information leakage of the training dataset." @default.
- W2969446490 created "2019-08-29" @default.
- W2969446490 creator A5012278873 @default.
- W2969446490 creator A5028791879 @default.
- W2969446490 creator A5040323946 @default.
- W2969446490 creator A5048139026 @default.
- W2969446490 creator A5052465353 @default.
- W2969446490 creator A5066056671 @default.
- W2969446490 creator A5071951805 @default.
- W2969446490 creator A5075729645 @default.
- W2969446490 date "2019-08-21" @default.
- W2969446490 modified "2023-09-22" @default.
- W2969446490 title "Generalization in Generative Adversarial Networks: A Novel Perspective from Privacy Protection" @default.
- W2969446490 cites W1886087434 @default.
- W2969446490 cites W2027595342 @default.
- W2969446490 cites W2099471712 @default.
- W2969446490 cites W2108598243 @default.
- W2969446490 cites W2504150216 @default.
- W2969446490 cites W2519536754 @default.
- W2969446490 cites W2535690855 @default.
- W2969446490 cites W2580360036 @default.
- W2969446490 cites W2593414223 @default.
- W2969446490 cites W2593729559 @default.
- W2969446490 cites W2605287558 @default.
- W2969446490 cites W2617174679 @default.
- W2969446490 cites W27434444 @default.
- W2969446490 cites W2780978202 @default.
- W2969446490 cites W2787248994 @default.
- W2969446490 cites W2795435272 @default.
- W2969446490 cites W2798729263 @default.
- W2969446490 cites W2904367110 @default.
- W2969446490 cites W2950893734 @default.
- W2969446490 cites W2962879692 @default.
- W2969446490 cites W2963373786 @default.
- W2969446490 cites W2963470893 @default.
- W2969446490 cites W2963684088 @default.
- W2969446490 cites W2963685250 @default.
- W2969446490 cites W2963800509 @default.
- W2969446490 cites W2963836885 @default.
- W2969446490 cites W2964121744 @default.
- W2969446490 cites W2964144352 @default.
- W2969446490 cites W2964237819 @default.
- W2969446490 cites W2964310974 @default.
- W2969446490 cites W2964326026 @default.
- W2969446490 cites W3118608800 @default.
- W2969446490 hasPublicationYear "2019" @default.
- W2969446490 type Work @default.
- W2969446490 sameAs 2969446490 @default.
- W2969446490 citedByCount "7" @default.
- W2969446490 countsByYear W29694464902019 @default.
- W2969446490 countsByYear W29694464902020 @default.
- W2969446490 countsByYear W29694464902021 @default.
- W2969446490 crossrefType "posted-content" @default.
- W2969446490 hasAuthorship W2969446490A5012278873 @default.
- W2969446490 hasAuthorship W2969446490A5028791879 @default.
- W2969446490 hasAuthorship W2969446490A5040323946 @default.
- W2969446490 hasAuthorship W2969446490A5048139026 @default.
- W2969446490 hasAuthorship W2969446490A5052465353 @default.
- W2969446490 hasAuthorship W2969446490A5066056671 @default.
- W2969446490 hasAuthorship W2969446490A5071951805 @default.
- W2969446490 hasAuthorship W2969446490A5075729645 @default.
- W2969446490 hasConcept C108583219 @default.
- W2969446490 hasConcept C119857082 @default.
- W2969446490 hasConcept C12713177 @default.
- W2969446490 hasConcept C134306372 @default.
- W2969446490 hasConcept C154945302 @default.
- W2969446490 hasConcept C177148314 @default.
- W2969446490 hasConcept C22019652 @default.
- W2969446490 hasConcept C2776135515 @default.
- W2969446490 hasConcept C2779201187 @default.
- W2969446490 hasConcept C2988773926 @default.
- W2969446490 hasConcept C33923547 @default.
- W2969446490 hasConcept C34388435 @default.
- W2969446490 hasConcept C37736160 @default.
- W2969446490 hasConcept C38652104 @default.
- W2969446490 hasConcept C39890363 @default.
- W2969446490 hasConcept C41008148 @default.
- W2969446490 hasConcept C50644808 @default.
- W2969446490 hasConcept C80444323 @default.
- W2969446490 hasConceptScore W2969446490C108583219 @default.
- W2969446490 hasConceptScore W2969446490C119857082 @default.
- W2969446490 hasConceptScore W2969446490C12713177 @default.
- W2969446490 hasConceptScore W2969446490C134306372 @default.
- W2969446490 hasConceptScore W2969446490C154945302 @default.
- W2969446490 hasConceptScore W2969446490C177148314 @default.
- W2969446490 hasConceptScore W2969446490C22019652 @default.
- W2969446490 hasConceptScore W2969446490C2776135515 @default.
- W2969446490 hasConceptScore W2969446490C2779201187 @default.
- W2969446490 hasConceptScore W2969446490C2988773926 @default.
- W2969446490 hasConceptScore W2969446490C33923547 @default.
- W2969446490 hasConceptScore W2969446490C34388435 @default.
- W2969446490 hasConceptScore W2969446490C37736160 @default.
- W2969446490 hasConceptScore W2969446490C38652104 @default.
- W2969446490 hasConceptScore W2969446490C39890363 @default.
- W2969446490 hasConceptScore W2969446490C41008148 @default.
- W2969446490 hasConceptScore W2969446490C50644808 @default.
- W2969446490 hasConceptScore W2969446490C80444323 @default.
- W2969446490 hasLocation W29694464901 @default.
- W2969446490 hasOpenAccess W2969446490 @default.
- W2969446490 hasPrimaryLocation W29694464901 @default.