Matches in SemOpenAlex for { <https://semopenalex.org/work/W4383221472> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4383221472 abstract "Many mobile applications have resorted to deep neural networks (DNNs) because of their strong inference capabilities. Since both input data and DNN architectures could be sensitive, there is an increasing demand for secure DNN execution on mobile devices. Towards this end, hardware-based trusted execution environments on mobile devices (mobile TEEs), such as ARM TrustZone, have recently been exploited to execute CNN securely. However, running entire DNNs on mobile TEEs is challenging as TEEs have stringent resource and performance constraints. In this work, we develop a novel mobile TEE-based security framework that can efficiently execute the entire DNN in a resource-constrained mobile TEE with minimal inference time overhead. Specifically, we propose a progressive pruning to gradually identify and remove the redundant neurons from a DNN while maintaining a high inference accuracy. Next, we develop a memory optimization method to deallocate the memory storage of the pruned neurons utilizing the low-level programming technique. Finally, we devise a novel adaptive partitioning method that divides the pruned model into multiple partitions according to the available memory in the mobile TEE and loads the partitions into the mobile TEE separately with a minimal loading time overhead. Our experiments with various DNNs and open-source datasets demonstrate that we can achieve 2-30 times less inference time with comparable accuracy compared to existing approaches securing entire DNNs with mobile TEE." @default.
- W4383221472 created "2023-07-06" @default.
- W4383221472 creator A5013956365 @default.
- W4383221472 creator A5016621512 @default.
- W4383221472 creator A5035193269 @default.
- W4383221472 creator A5062638134 @default.
- W4383221472 creator A5078477275 @default.
- W4383221472 creator A5080988193 @default.
- W4383221472 creator A5085616097 @default.
- W4383221472 date "2023-07-10" @default.
- W4383221472 modified "2023-09-26" @default.
- W4383221472 title "Secure and Efficient Mobile DNN Using Trusted Execution Environments" @default.
- W4383221472 cites W2097117768 @default.
- W4383221472 cites W2117539524 @default.
- W4383221472 cites W2137789775 @default.
- W4383221472 cites W2161030490 @default.
- W4383221472 cites W2161470918 @default.
- W4383221472 cites W2164327070 @default.
- W4383221472 cites W2518281301 @default.
- W4383221472 cites W2535690855 @default.
- W4383221472 cites W2591882872 @default.
- W4383221472 cites W2798170643 @default.
- W4383221472 cites W2886851211 @default.
- W4383221472 cites W2946026374 @default.
- W4383221472 cites W2962760690 @default.
- W4383221472 cites W2963456518 @default.
- W4383221472 cites W2963470893 @default.
- W4383221472 cites W2978814280 @default.
- W4383221472 cites W2997300524 @default.
- W4383221472 cites W2997768846 @default.
- W4383221472 cites W2998993395 @default.
- W4383221472 cites W3016075089 @default.
- W4383221472 cites W3016246341 @default.
- W4383221472 cites W3096785379 @default.
- W4383221472 cites W3099785009 @default.
- W4383221472 cites W3104849992 @default.
- W4383221472 doi "https://doi.org/10.1145/3579856.3582820" @default.
- W4383221472 hasPublicationYear "2023" @default.
- W4383221472 type Work @default.
- W4383221472 citedByCount "0" @default.
- W4383221472 crossrefType "proceedings-article" @default.
- W4383221472 hasAuthorship W4383221472A5013956365 @default.
- W4383221472 hasAuthorship W4383221472A5016621512 @default.
- W4383221472 hasAuthorship W4383221472A5035193269 @default.
- W4383221472 hasAuthorship W4383221472A5062638134 @default.
- W4383221472 hasAuthorship W4383221472A5078477275 @default.
- W4383221472 hasAuthorship W4383221472A5080988193 @default.
- W4383221472 hasAuthorship W4383221472A5085616097 @default.
- W4383221472 hasConcept C108010975 @default.
- W4383221472 hasConcept C111919701 @default.
- W4383221472 hasConcept C120314980 @default.
- W4383221472 hasConcept C144543869 @default.
- W4383221472 hasConcept C149635348 @default.
- W4383221472 hasConcept C154945302 @default.
- W4383221472 hasConcept C186967261 @default.
- W4383221472 hasConcept C2776214188 @default.
- W4383221472 hasConcept C2779960059 @default.
- W4383221472 hasConcept C2984842247 @default.
- W4383221472 hasConcept C31258907 @default.
- W4383221472 hasConcept C41008148 @default.
- W4383221472 hasConcept C50644808 @default.
- W4383221472 hasConcept C6557445 @default.
- W4383221472 hasConcept C79403827 @default.
- W4383221472 hasConcept C86803240 @default.
- W4383221472 hasConceptScore W4383221472C108010975 @default.
- W4383221472 hasConceptScore W4383221472C111919701 @default.
- W4383221472 hasConceptScore W4383221472C120314980 @default.
- W4383221472 hasConceptScore W4383221472C144543869 @default.
- W4383221472 hasConceptScore W4383221472C149635348 @default.
- W4383221472 hasConceptScore W4383221472C154945302 @default.
- W4383221472 hasConceptScore W4383221472C186967261 @default.
- W4383221472 hasConceptScore W4383221472C2776214188 @default.
- W4383221472 hasConceptScore W4383221472C2779960059 @default.
- W4383221472 hasConceptScore W4383221472C2984842247 @default.
- W4383221472 hasConceptScore W4383221472C31258907 @default.
- W4383221472 hasConceptScore W4383221472C41008148 @default.
- W4383221472 hasConceptScore W4383221472C50644808 @default.
- W4383221472 hasConceptScore W4383221472C6557445 @default.
- W4383221472 hasConceptScore W4383221472C79403827 @default.
- W4383221472 hasConceptScore W4383221472C86803240 @default.
- W4383221472 hasFunder F4320306076 @default.
- W4383221472 hasLocation W43832214721 @default.
- W4383221472 hasOpenAccess W4383221472 @default.
- W4383221472 hasPrimaryLocation W43832214721 @default.
- W4383221472 hasRelatedWork W123872086 @default.
- W4383221472 hasRelatedWork W2092071486 @default.
- W4383221472 hasRelatedWork W2131630752 @default.
- W4383221472 hasRelatedWork W2391167130 @default.
- W4383221472 hasRelatedWork W2785873404 @default.
- W4383221472 hasRelatedWork W2955715866 @default.
- W4383221472 hasRelatedWork W4283067488 @default.
- W4383221472 hasRelatedWork W94000989 @default.
- W4383221472 hasRelatedWork W2460246254 @default.
- W4383221472 hasRelatedWork W2516240846 @default.
- W4383221472 isParatext "false" @default.
- W4383221472 isRetracted "false" @default.
- W4383221472 workType "article" @default.