Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312795512> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W4312795512 endingPage "155" @default.
- W4312795512 startingPage "139" @default.
- W4312795512 abstract "Transfer learning through the use of pre-trained models has become a growing trend for the machine learning community. Consequently, numerous pre-trained models are released online to facilitate further research. However, it raises extensive concerns on whether these pre-trained models would leak privacy-sensitive information of their training data. Thus, in this work, we aim to answer the following questions: “Can we effectively recover private information from these pre-trained models? What are the sufficient conditions to retrieve such sensitive information?” We first explore different statistical information which can discriminate the private training distribution from other distributions. Based on our observations, we propose a novel private data reconstruction framework, SecretGen, to effectively recover private information. Compared with previous methods which can recover private data with the ground truth label of the targeted recovery instance, SecretGen does not require such prior knowledge, making it more practical. We conduct extensive experiments on different datasets under diverse scenarios to compare SecretGen with other baselines and provide a systematic benchmark to better understand the impact of different auxiliary information and optimization operations. We show that without prior knowledge about true class prediction, SecretGen is able to recover private data with similar performance compared with the ones that leverage such prior knowledge. If the prior knowledge is given, SecretGen will significantly outperform baseline methods. We also propose several quantitative metrics to further quantify the privacy vulnerability of pre-trained models, which will help the model selection for privacy-sensitive applications. Our code is available at: https://github.com/AI-secure/SecretGen ." @default.
- W4312795512 created "2023-01-05" @default.
- W4312795512 creator A5005843046 @default.
- W4312795512 creator A5022969296 @default.
- W4312795512 creator A5034045140 @default.
- W4312795512 creator A5034995105 @default.
- W4312795512 creator A5053064599 @default.
- W4312795512 date "2022-01-01" @default.
- W4312795512 modified "2023-10-17" @default.
- W4312795512 title "SecretGen: Privacy Recovery on Pre-trained Models via Distribution Discrimination" @default.
- W4312795512 cites W1834627138 @default.
- W4312795512 cites W2024922353 @default.
- W4312795512 cites W2051267297 @default.
- W4312795512 cites W2064076387 @default.
- W4312795512 cites W2108598243 @default.
- W4312795512 cites W2194775991 @default.
- W4312795512 cites W2515770085 @default.
- W4312795512 cites W2780958074 @default.
- W4312795512 cites W2963846044 @default.
- W4312795512 cites W3035616549 @default.
- W4312795512 cites W3096831136 @default.
- W4312795512 cites W3198429080 @default.
- W4312795512 cites W3212600502 @default.
- W4312795512 cites W4229820657 @default.
- W4312795512 doi "https://doi.org/10.1007/978-3-031-20065-6_9" @default.
- W4312795512 hasPublicationYear "2022" @default.
- W4312795512 type Work @default.
- W4312795512 citedByCount "0" @default.
- W4312795512 crossrefType "book-chapter" @default.
- W4312795512 hasAuthorship W4312795512A5005843046 @default.
- W4312795512 hasAuthorship W4312795512A5022969296 @default.
- W4312795512 hasAuthorship W4312795512A5034045140 @default.
- W4312795512 hasAuthorship W4312795512A5034995105 @default.
- W4312795512 hasAuthorship W4312795512A5053064599 @default.
- W4312795512 hasBestOaLocation W43127955122 @default.
- W4312795512 hasConcept C107673813 @default.
- W4312795512 hasConcept C111368507 @default.
- W4312795512 hasConcept C119857082 @default.
- W4312795512 hasConcept C123201435 @default.
- W4312795512 hasConcept C124101348 @default.
- W4312795512 hasConcept C12725497 @default.
- W4312795512 hasConcept C127313418 @default.
- W4312795512 hasConcept C13280743 @default.
- W4312795512 hasConcept C137822555 @default.
- W4312795512 hasConcept C146849305 @default.
- W4312795512 hasConcept C153083717 @default.
- W4312795512 hasConcept C154945302 @default.
- W4312795512 hasConcept C177769412 @default.
- W4312795512 hasConcept C185798385 @default.
- W4312795512 hasConcept C205649164 @default.
- W4312795512 hasConcept C38652104 @default.
- W4312795512 hasConcept C41008148 @default.
- W4312795512 hasConcept C99221444 @default.
- W4312795512 hasConceptScore W4312795512C107673813 @default.
- W4312795512 hasConceptScore W4312795512C111368507 @default.
- W4312795512 hasConceptScore W4312795512C119857082 @default.
- W4312795512 hasConceptScore W4312795512C123201435 @default.
- W4312795512 hasConceptScore W4312795512C124101348 @default.
- W4312795512 hasConceptScore W4312795512C12725497 @default.
- W4312795512 hasConceptScore W4312795512C127313418 @default.
- W4312795512 hasConceptScore W4312795512C13280743 @default.
- W4312795512 hasConceptScore W4312795512C137822555 @default.
- W4312795512 hasConceptScore W4312795512C146849305 @default.
- W4312795512 hasConceptScore W4312795512C153083717 @default.
- W4312795512 hasConceptScore W4312795512C154945302 @default.
- W4312795512 hasConceptScore W4312795512C177769412 @default.
- W4312795512 hasConceptScore W4312795512C185798385 @default.
- W4312795512 hasConceptScore W4312795512C205649164 @default.
- W4312795512 hasConceptScore W4312795512C38652104 @default.
- W4312795512 hasConceptScore W4312795512C41008148 @default.
- W4312795512 hasConceptScore W4312795512C99221444 @default.
- W4312795512 hasLocation W43127955121 @default.
- W4312795512 hasLocation W43127955122 @default.
- W4312795512 hasOpenAccess W4312795512 @default.
- W4312795512 hasPrimaryLocation W43127955121 @default.
- W4312795512 hasRelatedWork W2181911725 @default.
- W4312795512 hasRelatedWork W2262837473 @default.
- W4312795512 hasRelatedWork W2369625759 @default.
- W4312795512 hasRelatedWork W2594445182 @default.
- W4312795512 hasRelatedWork W2993584434 @default.
- W4312795512 hasRelatedWork W3048339612 @default.
- W4312795512 hasRelatedWork W3122870227 @default.
- W4312795512 hasRelatedWork W4288054996 @default.
- W4312795512 hasRelatedWork W4312795512 @default.
- W4312795512 hasRelatedWork W2571878527 @default.
- W4312795512 isParatext "false" @default.
- W4312795512 isRetracted "false" @default.
- W4312795512 workType "book-chapter" @default.