Matches in SemOpenAlex for { <https://semopenalex.org/work/W2903175372> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2903175372 endingPage "222" @default.
- W2903175372 startingPage "216" @default.
- W2903175372 abstract "Neural networks are becoming an attractive solution for automatizing vehicles in the automotive, military, and aerospace markets. Thanks to their low-cost, low-power consumption, and flexibility, field-programmable gate arrays (FPGAs) are among the promising devices to implement neural networks. Unfortunately, FPGAs are also known to be susceptible to radiation-induced errors. In this paper, we evaluate the effects of radiation-induced errors in the output correctness of two neural networks [Iris Flower artificial neural network (ANN) and Modified National Institute of Standards and Technology (MNIST) convolutional neural network (CNN)] implemented in static random-access memory-based FPGAs. In particular, we notice that radiation can induce errors that modify the output of the network with or without affecting the neural network’s functionality. We call the former critical errors and the latter tolerable errors. Through exhaustive fault injection, we identify the portions of Iris Flower ANN and MNIST CNN implementation on FPGAs that are more likely, once corrupted, to generate a critical or a tolerable error. Based on this analysis, we propose a selective hardening strategy that triplicates only the most vulnerable layers of the neural network. With neutron radiation testing, our selective hardening solution was able to mask 40% of faults with a marginal 8% overhead in one of our tested neural networks." @default.
- W2903175372 created "2018-12-11" @default.
- W2903175372 creator A5011899281 @default.
- W2903175372 creator A5021256073 @default.
- W2903175372 creator A5025490387 @default.
- W2903175372 creator A5041342112 @default.
- W2903175372 creator A5054264208 @default.
- W2903175372 creator A5070907101 @default.
- W2903175372 creator A5072470443 @default.
- W2903175372 date "2019-01-01" @default.
- W2903175372 modified "2023-10-05" @default.
- W2903175372 title "Selective Hardening for Neural Networks in FPGAs" @default.
- W2903175372 cites W1832437964 @default.
- W2903175372 cites W2001619934 @default.
- W2903175372 cites W2036190841 @default.
- W2903175372 cites W2061205755 @default.
- W2903175372 cites W2075547581 @default.
- W2903175372 cites W2102841035 @default.
- W2903175372 cites W2112796928 @default.
- W2903175372 cites W2146851106 @default.
- W2903175372 cites W2170171948 @default.
- W2903175372 cites W2464825514 @default.
- W2903175372 cites W2753775551 @default.
- W2903175372 cites W2767260595 @default.
- W2903175372 cites W2806983213 @default.
- W2903175372 cites W4252174618 @default.
- W2903175372 doi "https://doi.org/10.1109/tns.2018.2884460" @default.
- W2903175372 hasPublicationYear "2019" @default.
- W2903175372 type Work @default.
- W2903175372 sameAs 2903175372 @default.
- W2903175372 citedByCount "53" @default.
- W2903175372 countsByYear W29031753722019 @default.
- W2903175372 countsByYear W29031753722020 @default.
- W2903175372 countsByYear W29031753722021 @default.
- W2903175372 countsByYear W29031753722022 @default.
- W2903175372 countsByYear W29031753722023 @default.
- W2903175372 crossrefType "journal-article" @default.
- W2903175372 hasAuthorship W2903175372A5011899281 @default.
- W2903175372 hasAuthorship W2903175372A5021256073 @default.
- W2903175372 hasAuthorship W2903175372A5025490387 @default.
- W2903175372 hasAuthorship W2903175372A5041342112 @default.
- W2903175372 hasAuthorship W2903175372A5054264208 @default.
- W2903175372 hasAuthorship W2903175372A5070907101 @default.
- W2903175372 hasAuthorship W2903175372A5072470443 @default.
- W2903175372 hasConcept C120314980 @default.
- W2903175372 hasConcept C149635348 @default.
- W2903175372 hasConcept C154945302 @default.
- W2903175372 hasConcept C190502265 @default.
- W2903175372 hasConcept C41008148 @default.
- W2903175372 hasConcept C42935608 @default.
- W2903175372 hasConcept C50644808 @default.
- W2903175372 hasConcept C63540848 @default.
- W2903175372 hasConcept C81363708 @default.
- W2903175372 hasConceptScore W2903175372C120314980 @default.
- W2903175372 hasConceptScore W2903175372C149635348 @default.
- W2903175372 hasConceptScore W2903175372C154945302 @default.
- W2903175372 hasConceptScore W2903175372C190502265 @default.
- W2903175372 hasConceptScore W2903175372C41008148 @default.
- W2903175372 hasConceptScore W2903175372C42935608 @default.
- W2903175372 hasConceptScore W2903175372C50644808 @default.
- W2903175372 hasConceptScore W2903175372C63540848 @default.
- W2903175372 hasConceptScore W2903175372C81363708 @default.
- W2903175372 hasIssue "1" @default.
- W2903175372 hasLocation W29031753721 @default.
- W2903175372 hasOpenAccess W2903175372 @default.
- W2903175372 hasPrimaryLocation W29031753721 @default.
- W2903175372 hasRelatedWork W2435792570 @default.
- W2903175372 hasRelatedWork W2591882361 @default.
- W2903175372 hasRelatedWork W2978290780 @default.
- W2903175372 hasRelatedWork W2979891043 @default.
- W2903175372 hasRelatedWork W3012370783 @default.
- W2903175372 hasRelatedWork W3014927913 @default.
- W2903175372 hasRelatedWork W3108192785 @default.
- W2903175372 hasRelatedWork W3112665491 @default.
- W2903175372 hasRelatedWork W3127488929 @default.
- W2903175372 hasRelatedWork W4309224979 @default.
- W2903175372 hasVolume "66" @default.
- W2903175372 isParatext "false" @default.
- W2903175372 isRetracted "false" @default.
- W2903175372 magId "2903175372" @default.
- W2903175372 workType "article" @default.