Matches in SemOpenAlex for { <https://semopenalex.org/work/W3138090943> ?p ?o ?g. }
Showing items 1 to 58 of
58
with 100 items per page.
- W3138090943 abstract "In this paper, we address the problem of privacy-preserving training and evaluation of neural networks in an $N$-party, federated learning setting. We propose a novel system, POSEIDON, the first of its kind in the regime of privacy-preserving neural network training. It employs multiparty lattice-based cryptography to preserve the confidentiality of the training data, the model, and the evaluation data, under a passive-adversary model and collusions between up to $N-1$ parties. To efficiently execute the secure backpropagation algorithm for training neural networks, we provide a generic packing approach that enables Single Instruction, Multiple Data (SIMD) operations on encrypted data. We also introduce arbitrary linear transformations within the cryptographic bootstrapping operation, optimizing the costly cryptographic computations over the parties, and we define a constrained optimization problem for choosing the cryptographic parameters. Our experimental results show that POSEIDON achieves accuracy similar to centralized or decentralized non-private approaches and that its computation and communication overhead scales linearly with the number of parties. POSEIDON trains a 3-layer neural network on the MNIST dataset with 784 features and 60K samples distributed among 10 parties in less than 2 hours." @default.
- W3138090943 created "2021-03-29" @default.
- W3138090943 creator A5000583679 @default.
- W3138090943 creator A5025665736 @default.
- W3138090943 creator A5031281762 @default.
- W3138090943 creator A5072336168 @default.
- W3138090943 creator A5080497445 @default.
- W3138090943 creator A5089763199 @default.
- W3138090943 creator A5091369772 @default.
- W3138090943 date "2021-01-01" @default.
- W3138090943 modified "2023-10-16" @default.
- W3138090943 title "POSEIDON: Privacy-Preserving Federated Neural Network Learning" @default.
- W3138090943 doi "https://doi.org/10.14722/ndss.2021.24119" @default.
- W3138090943 hasPublicationYear "2021" @default.
- W3138090943 type Work @default.
- W3138090943 sameAs 3138090943 @default.
- W3138090943 citedByCount "32" @default.
- W3138090943 countsByYear W31380909432020 @default.
- W3138090943 countsByYear W31380909432021 @default.
- W3138090943 countsByYear W31380909432022 @default.
- W3138090943 countsByYear W31380909432023 @default.
- W3138090943 crossrefType "proceedings-article" @default.
- W3138090943 hasAuthorship W3138090943A5000583679 @default.
- W3138090943 hasAuthorship W3138090943A5025665736 @default.
- W3138090943 hasAuthorship W3138090943A5031281762 @default.
- W3138090943 hasAuthorship W3138090943A5072336168 @default.
- W3138090943 hasAuthorship W3138090943A5080497445 @default.
- W3138090943 hasAuthorship W3138090943A5089763199 @default.
- W3138090943 hasAuthorship W3138090943A5091369772 @default.
- W3138090943 hasBestOaLocation W31380909431 @default.
- W3138090943 hasConcept C123201435 @default.
- W3138090943 hasConcept C154945302 @default.
- W3138090943 hasConcept C38652104 @default.
- W3138090943 hasConcept C41008148 @default.
- W3138090943 hasConcept C50644808 @default.
- W3138090943 hasConceptScore W3138090943C123201435 @default.
- W3138090943 hasConceptScore W3138090943C154945302 @default.
- W3138090943 hasConceptScore W3138090943C38652104 @default.
- W3138090943 hasConceptScore W3138090943C41008148 @default.
- W3138090943 hasConceptScore W3138090943C50644808 @default.
- W3138090943 hasLocation W31380909431 @default.
- W3138090943 hasLocation W31380909432 @default.
- W3138090943 hasOpenAccess W3138090943 @default.
- W3138090943 hasPrimaryLocation W31380909431 @default.
- W3138090943 hasRelatedWork W2030496847 @default.
- W3138090943 hasRelatedWork W2067216284 @default.
- W3138090943 hasRelatedWork W2159443810 @default.
- W3138090943 hasRelatedWork W2386387936 @default.
- W3138090943 hasRelatedWork W2529433139 @default.
- W3138090943 hasRelatedWork W3001020386 @default.
- W3138090943 hasRelatedWork W3107474891 @default.
- W3138090943 hasRelatedWork W4362499384 @default.
- W3138090943 hasRelatedWork W644753246 @default.
- W3138090943 hasRelatedWork W1629725936 @default.
- W3138090943 isParatext "false" @default.
- W3138090943 isRetracted "false" @default.
- W3138090943 magId "3138090943" @default.
- W3138090943 workType "article" @default.