Matches in SemOpenAlex for { <https://semopenalex.org/work/W4221163902> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4221163902 abstract "Multiparty computation approaches to secure neural network inference traditionally rely on garbled circuits for securely executing nonlinear activation functions. However, garbled circuits require excessive communication between server and client, impose significant storage overheads, and incur large runtime penalties. To eliminate these costs, we propose an alternative to garbled circuits: Tabula, an algorithm based on secure lookup tables. Tabula leverages neural networks' ability to be quantized and employs a secure lookup table approach to efficiently, securely, and accurately compute neural network nonlinear activation functions. Compared to garbled circuits with quantized inputs, when computing individual nonlinear functions, our experiments show Tabula uses between $35 times$-$70 times$ less communication, is over $100times$ faster, and uses a comparable amount of storage. This leads to significant performance gains over garbled circuits with quantized inputs during secure inference on neural networks: Tabula reduces overall communication by up to $9 times$ and achieves a speedup of up to $50 times$, while imposing comparable storage costs." @default.
- W4221163902 created "2022-04-03" @default.
- W4221163902 creator A5000635267 @default.
- W4221163902 creator A5040197913 @default.
- W4221163902 creator A5043327132 @default.
- W4221163902 creator A5045336469 @default.
- W4221163902 creator A5062318856 @default.
- W4221163902 date "2022-03-05" @default.
- W4221163902 modified "2023-09-23" @default.
- W4221163902 title "Tabula: Efficiently Computing Nonlinear Activation Functions for Secure Neural Network Inference" @default.
- W4221163902 doi "https://doi.org/10.48550/arxiv.2203.02833" @default.
- W4221163902 hasPublicationYear "2022" @default.
- W4221163902 type Work @default.
- W4221163902 citedByCount "0" @default.
- W4221163902 crossrefType "posted-content" @default.
- W4221163902 hasAuthorship W4221163902A5000635267 @default.
- W4221163902 hasAuthorship W4221163902A5040197913 @default.
- W4221163902 hasAuthorship W4221163902A5043327132 @default.
- W4221163902 hasAuthorship W4221163902A5045336469 @default.
- W4221163902 hasAuthorship W4221163902A5062318856 @default.
- W4221163902 hasBestOaLocation W42211639021 @default.
- W4221163902 hasConcept C11413529 @default.
- W4221163902 hasConcept C119599485 @default.
- W4221163902 hasConcept C121332964 @default.
- W4221163902 hasConcept C124101348 @default.
- W4221163902 hasConcept C127413603 @default.
- W4221163902 hasConcept C134146338 @default.
- W4221163902 hasConcept C134835016 @default.
- W4221163902 hasConcept C154945302 @default.
- W4221163902 hasConcept C158622935 @default.
- W4221163902 hasConcept C173608175 @default.
- W4221163902 hasConcept C199360897 @default.
- W4221163902 hasConcept C2776214188 @default.
- W4221163902 hasConcept C41008148 @default.
- W4221163902 hasConcept C45235069 @default.
- W4221163902 hasConcept C45374587 @default.
- W4221163902 hasConcept C50644808 @default.
- W4221163902 hasConcept C62520636 @default.
- W4221163902 hasConcept C68339613 @default.
- W4221163902 hasConcept C80444323 @default.
- W4221163902 hasConceptScore W4221163902C11413529 @default.
- W4221163902 hasConceptScore W4221163902C119599485 @default.
- W4221163902 hasConceptScore W4221163902C121332964 @default.
- W4221163902 hasConceptScore W4221163902C124101348 @default.
- W4221163902 hasConceptScore W4221163902C127413603 @default.
- W4221163902 hasConceptScore W4221163902C134146338 @default.
- W4221163902 hasConceptScore W4221163902C134835016 @default.
- W4221163902 hasConceptScore W4221163902C154945302 @default.
- W4221163902 hasConceptScore W4221163902C158622935 @default.
- W4221163902 hasConceptScore W4221163902C173608175 @default.
- W4221163902 hasConceptScore W4221163902C199360897 @default.
- W4221163902 hasConceptScore W4221163902C2776214188 @default.
- W4221163902 hasConceptScore W4221163902C41008148 @default.
- W4221163902 hasConceptScore W4221163902C45235069 @default.
- W4221163902 hasConceptScore W4221163902C45374587 @default.
- W4221163902 hasConceptScore W4221163902C50644808 @default.
- W4221163902 hasConceptScore W4221163902C62520636 @default.
- W4221163902 hasConceptScore W4221163902C68339613 @default.
- W4221163902 hasConceptScore W4221163902C80444323 @default.
- W4221163902 hasLocation W42211639021 @default.
- W4221163902 hasOpenAccess W4221163902 @default.
- W4221163902 hasPrimaryLocation W42211639021 @default.
- W4221163902 hasRelatedWork W1978979050 @default.
- W4221163902 hasRelatedWork W1984116550 @default.
- W4221163902 hasRelatedWork W2001826080 @default.
- W4221163902 hasRelatedWork W2010976002 @default.
- W4221163902 hasRelatedWork W2170847085 @default.
- W4221163902 hasRelatedWork W2755467323 @default.
- W4221163902 hasRelatedWork W3040412425 @default.
- W4221163902 hasRelatedWork W3123269518 @default.
- W4221163902 hasRelatedWork W4200412787 @default.
- W4221163902 hasRelatedWork W744764827 @default.
- W4221163902 isParatext "false" @default.
- W4221163902 isRetracted "false" @default.
- W4221163902 workType "article" @default.