Matches in SemOpenAlex for { <https://semopenalex.org/work/W4286893974> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4286893974 abstract "Training high-performing deep learning models require a rich amount of data which is usually distributed among multiple data sources in practice. Simply centralizing these multi-sourced data for training would raise critical security and privacy concerns, and might be prohibited given the increasingly strict data regulations. To resolve the tension between privacy and data utilization in distributed learning, a machine learning framework called private aggregation of teacher ensembles(PATE) has been recently proposed. PATE harnesses the knowledge (label predictions for an unlabeled dataset) from distributed teacher models to train a student model, obviating access to distributed datasets. Despite being enticing, PATE does not offer protection for the individual label predictions from teacher models, which still entails privacy risks. In this paper, we propose SEDML, a new protocol which allows to securely and efficiently harness the distributed knowledge in machine learning. SEDML builds on lightweight cryptography and provides strong protection for the individual label predictions, as well as differential privacy guarantees on the aggregation results. Extensive evaluations show that while providing privacy protection, SEDML preserves the accuracy as in the plaintext baseline. Meanwhile, SEDML's performance in computing and communication is 43 times and 1.23 times higher than the latest technology, respectively." @default.
- W4286893974 created "2022-07-25" @default.
- W4286893974 creator A5036260603 @default.
- W4286893974 creator A5038015695 @default.
- W4286893974 creator A5061868304 @default.
- W4286893974 creator A5066887020 @default.
- W4286893974 creator A5067969908 @default.
- W4286893974 creator A5078158717 @default.
- W4286893974 date "2021-10-26" @default.
- W4286893974 modified "2023-10-18" @default.
- W4286893974 title "SEDML: Securely and Efficiently Harnessing Distributed Knowledge in Machine Learning" @default.
- W4286893974 doi "https://doi.org/10.48550/arxiv.2110.13499" @default.
- W4286893974 hasPublicationYear "2021" @default.
- W4286893974 type Work @default.
- W4286893974 citedByCount "0" @default.
- W4286893974 crossrefType "posted-content" @default.
- W4286893974 hasAuthorship W4286893974A5036260603 @default.
- W4286893974 hasAuthorship W4286893974A5038015695 @default.
- W4286893974 hasAuthorship W4286893974A5061868304 @default.
- W4286893974 hasAuthorship W4286893974A5066887020 @default.
- W4286893974 hasAuthorship W4286893974A5067969908 @default.
- W4286893974 hasAuthorship W4286893974A5078158717 @default.
- W4286893974 hasBestOaLocation W42868939741 @default.
- W4286893974 hasConcept C119857082 @default.
- W4286893974 hasConcept C123201435 @default.
- W4286893974 hasConcept C124101348 @default.
- W4286893974 hasConcept C142724271 @default.
- W4286893974 hasConcept C148730421 @default.
- W4286893974 hasConcept C154945302 @default.
- W4286893974 hasConcept C15744967 @default.
- W4286893974 hasConcept C178489894 @default.
- W4286893974 hasConcept C180706569 @default.
- W4286893974 hasConcept C19417346 @default.
- W4286893974 hasConcept C204787440 @default.
- W4286893974 hasConcept C23130292 @default.
- W4286893974 hasConcept C2522767166 @default.
- W4286893974 hasConcept C2779582901 @default.
- W4286893974 hasConcept C2780385302 @default.
- W4286893974 hasConcept C2992525071 @default.
- W4286893974 hasConcept C33884865 @default.
- W4286893974 hasConcept C38652104 @default.
- W4286893974 hasConcept C41008148 @default.
- W4286893974 hasConcept C71924100 @default.
- W4286893974 hasConcept C92717368 @default.
- W4286893974 hasConceptScore W4286893974C119857082 @default.
- W4286893974 hasConceptScore W4286893974C123201435 @default.
- W4286893974 hasConceptScore W4286893974C124101348 @default.
- W4286893974 hasConceptScore W4286893974C142724271 @default.
- W4286893974 hasConceptScore W4286893974C148730421 @default.
- W4286893974 hasConceptScore W4286893974C154945302 @default.
- W4286893974 hasConceptScore W4286893974C15744967 @default.
- W4286893974 hasConceptScore W4286893974C178489894 @default.
- W4286893974 hasConceptScore W4286893974C180706569 @default.
- W4286893974 hasConceptScore W4286893974C19417346 @default.
- W4286893974 hasConceptScore W4286893974C204787440 @default.
- W4286893974 hasConceptScore W4286893974C23130292 @default.
- W4286893974 hasConceptScore W4286893974C2522767166 @default.
- W4286893974 hasConceptScore W4286893974C2779582901 @default.
- W4286893974 hasConceptScore W4286893974C2780385302 @default.
- W4286893974 hasConceptScore W4286893974C2992525071 @default.
- W4286893974 hasConceptScore W4286893974C33884865 @default.
- W4286893974 hasConceptScore W4286893974C38652104 @default.
- W4286893974 hasConceptScore W4286893974C41008148 @default.
- W4286893974 hasConceptScore W4286893974C71924100 @default.
- W4286893974 hasConceptScore W4286893974C92717368 @default.
- W4286893974 hasLocation W42868939741 @default.
- W4286893974 hasOpenAccess W4286893974 @default.
- W4286893974 hasPrimaryLocation W42868939741 @default.
- W4286893974 hasRelatedWork W2148742293 @default.
- W4286893974 hasRelatedWork W2352858694 @default.
- W4286893974 hasRelatedWork W2774387311 @default.
- W4286893974 hasRelatedWork W4200630816 @default.
- W4286893974 hasRelatedWork W4221146017 @default.
- W4286893974 hasRelatedWork W4226109007 @default.
- W4286893974 hasRelatedWork W4286893974 @default.
- W4286893974 hasRelatedWork W4287812080 @default.
- W4286893974 hasRelatedWork W4360993981 @default.
- W4286893974 hasRelatedWork W2184964202 @default.
- W4286893974 isParatext "false" @default.
- W4286893974 isRetracted "false" @default.
- W4286893974 workType "article" @default.