Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288796757> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4288796757 abstract "Nowadays, machine learning models, especially neural networks, become prevalent in many real-world applications.These models are trained based on a one-way trip from user data: as long as users contribute their data, there is no way to withdraw; and it is well-known that a neural network memorizes its training data. This contradicts the right to be forgotten clause of GDPR, potentially leading to law violations. To this end, machine unlearning becomes a popular research topic, which allows users to eliminate memorization of their private data from a trained machine learning model.In this paper, we propose the first uniform metric called for-getting rate to measure the effectiveness of a machine unlearning method. It is based on the concept of membership inference and describes the transformation rate of the eliminated data from memorized to unknown after conducting unlearning. We also propose a novel unlearning method calledForsaken. It is superior to previous work in either utility or efficiency (when achieving the same forgetting rate). We benchmark Forsaken with eight standard datasets to evaluate its performance. The experimental results show that it can achieve more than 90% forgetting rate on average and only causeless than 5% accuracy loss." @default.
- W4288796757 created "2022-07-30" @default.
- W4288796757 creator A5012016098 @default.
- W4288796757 creator A5014861105 @default.
- W4288796757 creator A5023363049 @default.
- W4288796757 creator A5036357902 @default.
- W4288796757 creator A5056982659 @default.
- W4288796757 creator A5058120371 @default.
- W4288796757 creator A5060165452 @default.
- W4288796757 creator A5087996736 @default.
- W4288796757 date "2020-03-24" @default.
- W4288796757 modified "2023-09-25" @default.
- W4288796757 title "Learn to Forget: Machine Unlearning via Neuron Masking" @default.
- W4288796757 doi "https://doi.org/10.48550/arxiv.2003.10933" @default.
- W4288796757 hasPublicationYear "2020" @default.
- W4288796757 type Work @default.
- W4288796757 citedByCount "0" @default.
- W4288796757 crossrefType "posted-content" @default.
- W4288796757 hasAuthorship W4288796757A5012016098 @default.
- W4288796757 hasAuthorship W4288796757A5014861105 @default.
- W4288796757 hasAuthorship W4288796757A5023363049 @default.
- W4288796757 hasAuthorship W4288796757A5036357902 @default.
- W4288796757 hasAuthorship W4288796757A5056982659 @default.
- W4288796757 hasAuthorship W4288796757A5058120371 @default.
- W4288796757 hasAuthorship W4288796757A5060165452 @default.
- W4288796757 hasAuthorship W4288796757A5087996736 @default.
- W4288796757 hasBestOaLocation W42887967571 @default.
- W4288796757 hasConcept C104317684 @default.
- W4288796757 hasConcept C119857082 @default.
- W4288796757 hasConcept C127413603 @default.
- W4288796757 hasConcept C13280743 @default.
- W4288796757 hasConcept C145420912 @default.
- W4288796757 hasConcept C154945302 @default.
- W4288796757 hasConcept C15744967 @default.
- W4288796757 hasConcept C176217482 @default.
- W4288796757 hasConcept C180747234 @default.
- W4288796757 hasConcept C185592680 @default.
- W4288796757 hasConcept C185798385 @default.
- W4288796757 hasConcept C204241405 @default.
- W4288796757 hasConcept C205649164 @default.
- W4288796757 hasConcept C21547014 @default.
- W4288796757 hasConcept C2776214188 @default.
- W4288796757 hasConcept C30038468 @default.
- W4288796757 hasConcept C41008148 @default.
- W4288796757 hasConcept C50644808 @default.
- W4288796757 hasConcept C55493867 @default.
- W4288796757 hasConcept C7149132 @default.
- W4288796757 hasConceptScore W4288796757C104317684 @default.
- W4288796757 hasConceptScore W4288796757C119857082 @default.
- W4288796757 hasConceptScore W4288796757C127413603 @default.
- W4288796757 hasConceptScore W4288796757C13280743 @default.
- W4288796757 hasConceptScore W4288796757C145420912 @default.
- W4288796757 hasConceptScore W4288796757C154945302 @default.
- W4288796757 hasConceptScore W4288796757C15744967 @default.
- W4288796757 hasConceptScore W4288796757C176217482 @default.
- W4288796757 hasConceptScore W4288796757C180747234 @default.
- W4288796757 hasConceptScore W4288796757C185592680 @default.
- W4288796757 hasConceptScore W4288796757C185798385 @default.
- W4288796757 hasConceptScore W4288796757C204241405 @default.
- W4288796757 hasConceptScore W4288796757C205649164 @default.
- W4288796757 hasConceptScore W4288796757C21547014 @default.
- W4288796757 hasConceptScore W4288796757C2776214188 @default.
- W4288796757 hasConceptScore W4288796757C30038468 @default.
- W4288796757 hasConceptScore W4288796757C41008148 @default.
- W4288796757 hasConceptScore W4288796757C50644808 @default.
- W4288796757 hasConceptScore W4288796757C55493867 @default.
- W4288796757 hasConceptScore W4288796757C7149132 @default.
- W4288796757 hasLocation W42887967571 @default.
- W4288796757 hasOpenAccess W4288796757 @default.
- W4288796757 hasPrimaryLocation W42887967571 @default.
- W4288796757 hasRelatedWork W10389550 @default.
- W4288796757 hasRelatedWork W10502778 @default.
- W4288796757 hasRelatedWork W11356396 @default.
- W4288796757 hasRelatedWork W14986123 @default.
- W4288796757 hasRelatedWork W14993799 @default.
- W4288796757 hasRelatedWork W2701911 @default.
- W4288796757 hasRelatedWork W3370582 @default.
- W4288796757 hasRelatedWork W482614 @default.
- W4288796757 hasRelatedWork W7339520 @default.
- W4288796757 hasRelatedWork W8021486 @default.
- W4288796757 isParatext "false" @default.
- W4288796757 isRetracted "false" @default.
- W4288796757 workType "article" @default.