Matches in SemOpenAlex for { <https://semopenalex.org/work/W4200565739> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W4200565739 endingPage "107850" @default.
- W4200565739 startingPage "107850" @default.
- W4200565739 abstract "Multi-party learning provides effective solutions for building a jointly-trained model with scattered data while meeting user privacy, data security and government regulations. Nevertheless, since the knowledge sharing process among multiple participants is conducted by uploading model parameters instead of individual private data, local poisoning attacks in the multi-party setting could be more covert and destructive. In this paper, we propose a novel immune system deployed in the multi-party learning scenario (MPIS) to defend against local poisoning attacks. We investigate the commonality between the biological immunity and the defense against poisoning attacks, and analogize the secure defense framework as an immune system. The approach achieves antigen recognition, immune response and immunological memory with an adversarial pipeline, which is not limited by the number of compromised clients or the duration of their involvement, and it is capable of adaptively determine the aggregation weight of different local models. Extensive experimental results on image and text datasets with different neural networks demonstrate the superiority of the MPIS framework in both model performance and robustness against poisoning attacks." @default.
- W4200565739 created "2021-12-31" @default.
- W4200565739 creator A5011958968 @default.
- W4200565739 creator A5038907917 @default.
- W4200565739 creator A5066877556 @default.
- W4200565739 creator A5090116897 @default.
- W4200565739 date "2022-02-01" @default.
- W4200565739 modified "2023-10-16" @default.
- W4200565739 title "Defending local poisoning attacks in multi-party learning via immune system" @default.
- W4200565739 cites W1965607176 @default.
- W4200565739 cites W1982141954 @default.
- W4200565739 cites W2007339694 @default.
- W4200565739 cites W2096299607 @default.
- W4200565739 cites W2102785645 @default.
- W4200565739 cites W2104266972 @default.
- W4200565739 cites W2255388164 @default.
- W4200565739 cites W2321189130 @default.
- W4200565739 cites W2555390854 @default.
- W4200565739 cites W2782073709 @default.
- W4200565739 cites W2805151692 @default.
- W4200565739 cites W2887063112 @default.
- W4200565739 cites W2905552164 @default.
- W4200565739 cites W2962763344 @default.
- W4200565739 cites W2963858765 @default.
- W4200565739 cites W2975050769 @default.
- W4200565739 cites W2993953869 @default.
- W4200565739 cites W2997357271 @default.
- W4200565739 cites W3004155269 @default.
- W4200565739 cites W3034912136 @default.
- W4200565739 cites W3037886022 @default.
- W4200565739 cites W3093177164 @default.
- W4200565739 cites W3123459983 @default.
- W4200565739 doi "https://doi.org/10.1016/j.knosys.2021.107850" @default.
- W4200565739 hasPublicationYear "2022" @default.
- W4200565739 type Work @default.
- W4200565739 citedByCount "3" @default.
- W4200565739 countsByYear W42005657392022 @default.
- W4200565739 crossrefType "journal-article" @default.
- W4200565739 hasAuthorship W4200565739A5011958968 @default.
- W4200565739 hasAuthorship W4200565739A5038907917 @default.
- W4200565739 hasAuthorship W4200565739A5066877556 @default.
- W4200565739 hasAuthorship W4200565739A5090116897 @default.
- W4200565739 hasConcept C104317684 @default.
- W4200565739 hasConcept C119857082 @default.
- W4200565739 hasConcept C136764020 @default.
- W4200565739 hasConcept C138885662 @default.
- W4200565739 hasConcept C154945302 @default.
- W4200565739 hasConcept C185592680 @default.
- W4200565739 hasConcept C2779338814 @default.
- W4200565739 hasConcept C38652104 @default.
- W4200565739 hasConcept C41008148 @default.
- W4200565739 hasConcept C41895202 @default.
- W4200565739 hasConcept C55493867 @default.
- W4200565739 hasConcept C63479239 @default.
- W4200565739 hasConcept C71901391 @default.
- W4200565739 hasConceptScore W4200565739C104317684 @default.
- W4200565739 hasConceptScore W4200565739C119857082 @default.
- W4200565739 hasConceptScore W4200565739C136764020 @default.
- W4200565739 hasConceptScore W4200565739C138885662 @default.
- W4200565739 hasConceptScore W4200565739C154945302 @default.
- W4200565739 hasConceptScore W4200565739C185592680 @default.
- W4200565739 hasConceptScore W4200565739C2779338814 @default.
- W4200565739 hasConceptScore W4200565739C38652104 @default.
- W4200565739 hasConceptScore W4200565739C41008148 @default.
- W4200565739 hasConceptScore W4200565739C41895202 @default.
- W4200565739 hasConceptScore W4200565739C55493867 @default.
- W4200565739 hasConceptScore W4200565739C63479239 @default.
- W4200565739 hasConceptScore W4200565739C71901391 @default.
- W4200565739 hasLocation W42005657391 @default.
- W4200565739 hasOpenAccess W4200565739 @default.
- W4200565739 hasPrimaryLocation W42005657391 @default.
- W4200565739 hasRelatedWork W1983811306 @default.
- W4200565739 hasRelatedWork W2089848187 @default.
- W4200565739 hasRelatedWork W2358742051 @default.
- W4200565739 hasRelatedWork W2365093105 @default.
- W4200565739 hasRelatedWork W2516405122 @default.
- W4200565739 hasRelatedWork W2961085424 @default.
- W4200565739 hasRelatedWork W4286629047 @default.
- W4200565739 hasRelatedWork W4306321456 @default.
- W4200565739 hasRelatedWork W4306674287 @default.
- W4200565739 hasRelatedWork W4224009465 @default.
- W4200565739 hasVolume "238" @default.
- W4200565739 isParatext "false" @default.
- W4200565739 isRetracted "false" @default.
- W4200565739 workType "article" @default.