Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048665067> ?p ?o ?g. }
- W3048665067 abstract "Today's cloud vendors are competing to provide various offerings to simplify and accelerate AI service deployment. However, cloud users always have concerns about the confidentiality of their runtime data, which are supposed to be processed on third-party's compute infrastructures. Information disclosure of user-supplied data may jeopardize users' privacy and breach increasingly stringent data protection regulations. In this paper, we systematically investigate the life cycles of inference inputs in deep learning image classification pipelines and understand how the information could be leaked. Based on the discovered insights, we develop a Ternary Model Partitioning mechanism and bring trusted execution environments to mitigate the identified information leakages. Our research prototype consists of two co-operative components: (1) Model Assessment Framework, a local model evaluation and partitioning tool that assists cloud users in deployment preparation; (2) Infenclave, an enclave-based model serving system for online confidential inference in the cloud. We have conducted comprehensive security and performance evaluation on three representative ImageNet-level deep learning models with different network depths and architectural complexity. Our results demonstrate the feasibility of launching confidential inference services in the cloud with maximized confidentiality guarantees and low performance costs." @default.
- W3048665067 created "2020-08-18" @default.
- W3048665067 creator A5005966189 @default.
- W3048665067 creator A5006668000 @default.
- W3048665067 creator A5033144884 @default.
- W3048665067 creator A5037005556 @default.
- W3048665067 creator A5037838916 @default.
- W3048665067 creator A5040348286 @default.
- W3048665067 creator A5057478629 @default.
- W3048665067 creator A5071982602 @default.
- W3048665067 date "2018-07-03" @default.
- W3048665067 modified "2023-09-27" @default.
- W3048665067 title "Confidential Inference via Ternary Model Partitioning" @default.
- W3048665067 cites W1510952750 @default.
- W3048665067 cites W1528076390 @default.
- W3048665067 cites W1533861849 @default.
- W3048665067 cites W1787224781 @default.
- W3048665067 cites W1825675169 @default.
- W3048665067 cites W1849277567 @default.
- W3048665067 cites W1915485278 @default.
- W3048665067 cites W2009733253 @default.
- W3048665067 cites W2036329595 @default.
- W3048665067 cites W2053637704 @default.
- W3048665067 cites W2093367651 @default.
- W3048665067 cites W2106463421 @default.
- W3048665067 cites W2107878631 @default.
- W3048665067 cites W2123045220 @default.
- W3048665067 cites W2163605009 @default.
- W3048665067 cites W2295107390 @default.
- W3048665067 cites W2397423248 @default.
- W3048665067 cites W2435473771 @default.
- W3048665067 cites W2460441129 @default.
- W3048665067 cites W2484015479 @default.
- W3048665067 cites W2541884796 @default.
- W3048665067 cites W2576037784 @default.
- W3048665067 cites W2593634001 @default.
- W3048665067 cites W2701059868 @default.
- W3048665067 cites W2754620673 @default.
- W3048665067 cites W2765200655 @default.
- W3048665067 cites W2788636062 @default.
- W3048665067 cites W2793398195 @default.
- W3048665067 cites W2805074088 @default.
- W3048665067 cites W2885216147 @default.
- W3048665067 cites W2886576854 @default.
- W3048665067 cites W2888798936 @default.
- W3048665067 cites W2945838820 @default.
- W3048665067 cites W2954241526 @default.
- W3048665067 cites W2962851944 @default.
- W3048665067 cites W2962858109 @default.
- W3048665067 cites W2963446712 @default.
- W3048665067 cites W2963989815 @default.
- W3048665067 cites W2969460181 @default.
- W3048665067 hasPublicationYear "2018" @default.
- W3048665067 type Work @default.
- W3048665067 sameAs 3048665067 @default.
- W3048665067 citedByCount "1" @default.
- W3048665067 countsByYear W30486650672021 @default.
- W3048665067 crossrefType "posted-content" @default.
- W3048665067 hasAuthorship W3048665067A5005966189 @default.
- W3048665067 hasAuthorship W3048665067A5006668000 @default.
- W3048665067 hasAuthorship W3048665067A5033144884 @default.
- W3048665067 hasAuthorship W3048665067A5037005556 @default.
- W3048665067 hasAuthorship W3048665067A5037838916 @default.
- W3048665067 hasAuthorship W3048665067A5040348286 @default.
- W3048665067 hasAuthorship W3048665067A5057478629 @default.
- W3048665067 hasAuthorship W3048665067A5071982602 @default.
- W3048665067 hasConcept C105339364 @default.
- W3048665067 hasConcept C108583219 @default.
- W3048665067 hasConcept C111919701 @default.
- W3048665067 hasConcept C115903868 @default.
- W3048665067 hasConcept C154945302 @default.
- W3048665067 hasConcept C2776214188 @default.
- W3048665067 hasConcept C38652104 @default.
- W3048665067 hasConcept C41008148 @default.
- W3048665067 hasConcept C71745522 @default.
- W3048665067 hasConcept C79974875 @default.
- W3048665067 hasConceptScore W3048665067C105339364 @default.
- W3048665067 hasConceptScore W3048665067C108583219 @default.
- W3048665067 hasConceptScore W3048665067C111919701 @default.
- W3048665067 hasConceptScore W3048665067C115903868 @default.
- W3048665067 hasConceptScore W3048665067C154945302 @default.
- W3048665067 hasConceptScore W3048665067C2776214188 @default.
- W3048665067 hasConceptScore W3048665067C38652104 @default.
- W3048665067 hasConceptScore W3048665067C41008148 @default.
- W3048665067 hasConceptScore W3048665067C71745522 @default.
- W3048665067 hasConceptScore W3048665067C79974875 @default.
- W3048665067 hasLocation W30486650671 @default.
- W3048665067 hasOpenAccess W3048665067 @default.
- W3048665067 hasPrimaryLocation W30486650671 @default.
- W3048665067 hasRelatedWork W1973422254 @default.
- W3048665067 hasRelatedWork W2134600520 @default.
- W3048665067 hasRelatedWork W2166313909 @default.
- W3048665067 hasRelatedWork W2185642343 @default.
- W3048665067 hasRelatedWork W2186035926 @default.
- W3048665067 hasRelatedWork W2189374852 @default.
- W3048665067 hasRelatedWork W2242336816 @default.
- W3048665067 hasRelatedWork W2290479004 @default.
- W3048665067 hasRelatedWork W2417576116 @default.
- W3048665067 hasRelatedWork W2474661953 @default.
- W3048665067 hasRelatedWork W2562432505 @default.