Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136251605> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3136251605 endingPage "708" @default.
- W3136251605 startingPage "699" @default.
- W3136251605 abstract " Deep Neural Networks (DNNs) used in safetycritical systems cannot compromise their performance due to reliability issues. In particular, soft errors are the worst. Selective softwarebased protection solutions are among the best techniques to improve the reliability of DNNs efficiently. However, their most significant challenge is precisely hardening portions of the DNN model to avoid performance degradation. In this work, we propose a comprehensive methodology to analyze the reliability of object detection and classification algorithms run on GPUs from the lowest (instruction) evaluation level. The ultimate goal is to avoid the performance penalty of full instruction duplication by confidently identifying the vulnerable instructions. For this purpose, we propose a technique, Instruction Vulnerability Factor (IVF). By applying our methodology on ResNet and YOLO models, we demonstrate that both models’ most vulnerable instructions can be precisely determined. Moreover, we show that YOLO is more sensitive to the changes caused by soft errors than ResNet. Also, ResNet depends on the input image in its reliability, while YOLO tends to be independent." @default.
- W3136251605 created "2021-03-29" @default.
- W3136251605 creator A5022944277 @default.
- W3136251605 creator A5053572855 @default.
- W3136251605 creator A5056484501 @default.
- W3136251605 creator A5059117888 @default.
- W3136251605 creator A5071458554 @default.
- W3136251605 creator A5071984203 @default.
- W3136251605 date "2020-12-08" @default.
- W3136251605 modified "2023-10-16" @default.
- W3136251605 title "Analyzing the Impact of Soft Errors in Deep Neural Networks on GPUsfrom Instruction Level" @default.
- W3136251605 cites W2097117768 @default.
- W3136251605 cites W2098290747 @default.
- W3136251605 cites W2618530766 @default.
- W3136251605 cites W2735162286 @default.
- W3136251605 cites W2765567583 @default.
- W3136251605 cites W2884175902 @default.
- W3136251605 cites W2884367402 @default.
- W3136251605 cites W2901073342 @default.
- W3136251605 cites W2901848761 @default.
- W3136251605 cites W2902644322 @default.
- W3136251605 cites W2959716684 @default.
- W3136251605 cites W2972006294 @default.
- W3136251605 cites W2983012040 @default.
- W3136251605 cites W3002446690 @default.
- W3136251605 cites W3007493000 @default.
- W3136251605 cites W3017855299 @default.
- W3136251605 cites W3108379332 @default.
- W3136251605 cites W4244505835 @default.
- W3136251605 doi "https://doi.org/10.37394/23203.2020.15.70" @default.
- W3136251605 hasPublicationYear "2020" @default.
- W3136251605 type Work @default.
- W3136251605 sameAs 3136251605 @default.
- W3136251605 citedByCount "5" @default.
- W3136251605 countsByYear W31362516052021 @default.
- W3136251605 countsByYear W31362516052022 @default.
- W3136251605 crossrefType "journal-article" @default.
- W3136251605 hasAuthorship W3136251605A5022944277 @default.
- W3136251605 hasAuthorship W3136251605A5053572855 @default.
- W3136251605 hasAuthorship W3136251605A5056484501 @default.
- W3136251605 hasAuthorship W3136251605A5059117888 @default.
- W3136251605 hasAuthorship W3136251605A5071458554 @default.
- W3136251605 hasAuthorship W3136251605A5071984203 @default.
- W3136251605 hasBestOaLocation W31362516051 @default.
- W3136251605 hasConcept C119857082 @default.
- W3136251605 hasConcept C121332964 @default.
- W3136251605 hasConcept C127413603 @default.
- W3136251605 hasConcept C154945302 @default.
- W3136251605 hasConcept C163258240 @default.
- W3136251605 hasConcept C200601418 @default.
- W3136251605 hasConcept C2944601119 @default.
- W3136251605 hasConcept C2984842247 @default.
- W3136251605 hasConcept C38652104 @default.
- W3136251605 hasConcept C41008148 @default.
- W3136251605 hasConcept C43214815 @default.
- W3136251605 hasConcept C50644808 @default.
- W3136251605 hasConcept C62520636 @default.
- W3136251605 hasConcept C95713431 @default.
- W3136251605 hasConceptScore W3136251605C119857082 @default.
- W3136251605 hasConceptScore W3136251605C121332964 @default.
- W3136251605 hasConceptScore W3136251605C127413603 @default.
- W3136251605 hasConceptScore W3136251605C154945302 @default.
- W3136251605 hasConceptScore W3136251605C163258240 @default.
- W3136251605 hasConceptScore W3136251605C200601418 @default.
- W3136251605 hasConceptScore W3136251605C2944601119 @default.
- W3136251605 hasConceptScore W3136251605C2984842247 @default.
- W3136251605 hasConceptScore W3136251605C38652104 @default.
- W3136251605 hasConceptScore W3136251605C41008148 @default.
- W3136251605 hasConceptScore W3136251605C43214815 @default.
- W3136251605 hasConceptScore W3136251605C50644808 @default.
- W3136251605 hasConceptScore W3136251605C62520636 @default.
- W3136251605 hasConceptScore W3136251605C95713431 @default.
- W3136251605 hasLocation W31362516051 @default.
- W3136251605 hasOpenAccess W3136251605 @default.
- W3136251605 hasPrimaryLocation W31362516051 @default.
- W3136251605 hasRelatedWork W2077367045 @default.
- W3136251605 hasRelatedWork W2905637214 @default.
- W3136251605 hasRelatedWork W2950066684 @default.
- W3136251605 hasRelatedWork W3048339032 @default.
- W3136251605 hasRelatedWork W4288853838 @default.
- W3136251605 hasRelatedWork W4298388782 @default.
- W3136251605 hasRelatedWork W4320854192 @default.
- W3136251605 hasRelatedWork W4322008322 @default.
- W3136251605 hasRelatedWork W4324297365 @default.
- W3136251605 hasRelatedWork W1629725936 @default.
- W3136251605 hasVolume "15" @default.
- W3136251605 isParatext "false" @default.
- W3136251605 isRetracted "false" @default.
- W3136251605 magId "3136251605" @default.
- W3136251605 workType "article" @default.