FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3161130147> ?p ?o ?g. }

• W3161130147 abstract "Deep neural networks (DNNs) are becoming an integral part of most software systems. Previous work has shown that DNNs have bugs. Unfortunately, existing debugging techniques do not support localizing DNN bugs because of the lack of understanding of model behaviors. The entire DNN model appears as a black box. To address these problems, we propose an approach that automatically determines whether the model is buggy or not, and identifies the root causes. Our key insight is that historic trends in values propagated between layers can be analyzed to identify faults, and localize faults. To that end, we first enable dynamic analysis of deep learning applications: by converting it into an imperative representation and alternatively using a callback mechanism. Both mechanisms allows us to insert probes that enable dynamic analysis over the traces produced by the DNN while it is being trained on the training data. We then conduct dynamic analysis over the traces to identify the faulty layer that causes the error. We propose an algorithm for identifying root causes by capturing any numerical error and monitoring the model during training and finding the relevance of every layer on the DNN outcome. We have collected a benchmark containing 40 buggy models and patches that contain real errors in deep learning applications from Stack Overflow and GitHub. Our benchmark can be used to evaluate automated debugging tools and repair techniques. We have evaluated our approach using this DNN bug-and-patch benchmark, and the results showed that our approach is much more effective than the existing debugging approach used in the state of the practice Keras library. For 34 out of 40 cases, our approach was able to detect faults whereas the best debugging approach provided by Keras detected 32 out of 40 faults. Our approach was able to localize 21 out of 40 bugs whereas Keras did not localize any faults." @default.
• W3161130147 created "2021-05-24" @default.
• W3161130147 creator A5004183493 @default.
• W3161130147 creator A5059626072 @default.
• W3161130147 creator A5074426991 @default.
• W3161130147 date "2021-03-04" @default.
• W3161130147 modified "2023-09-27" @default.
• W3161130147 title "DeepLocalize: Fault Localization for Deep Neural Networks" @default.
• W3161130147 cites W2016027000 @default.
• W3161130147 cites W2092239677 @default.
• W3161130147 cites W2121081915 @default.
• W3161130147 cites W2133671888 @default.
• W3161130147 cites W2145373440 @default.
• W3161130147 cites W2148329403 @default.
• W3161130147 cites W2155893237 @default.
• W3161130147 cites W2156357889 @default.
• W3161130147 cites W2162045655 @default.
• W3161130147 cites W2166007208 @default.
• W3161130147 cites W2168561184 @default.
• W3161130147 cites W2186615578 @default.
• W3161130147 cites W2384495648 @default.
• W3161130147 cites W2523246573 @default.
• W3161130147 cites W2786076697 @default.
• W3161130147 cites W2850992922 @default.
• W3161130147 cites W2898868990 @default.
• W3161130147 cites W2899771611 @default.
• W3161130147 cites W2912440308 @default.
• W3161130147 cites W2954252370 @default.
• W3161130147 cites W2963327228 @default.
• W3161130147 cites W2963446085 @default.
• W3161130147 cites W2968594320 @default.
• W3161130147 cites W2996489182 @default.
• W3161130147 cites W2999691867 @default.
• W3161130147 cites W3005940936 @default.
• W3161130147 cites W3006235430 @default.
• W3161130147 cites W3007157104 @default.
• W3161130147 cites W3091633490 @default.
• W3161130147 cites W3103697033 @default.
• W3161130147 hasPublicationYear "2021" @default.
• W3161130147 type Work @default.
• W3161130147 sameAs 3161130147 @default.
• W3161130147 citedByCount "0" @default.
• W3161130147 crossrefType "posted-content" @default.
• W3161130147 hasAuthorship W3161130147A5004183493 @default.
• W3161130147 hasAuthorship W3161130147A5059626072 @default.
• W3161130147 hasAuthorship W3161130147A5074426991 @default.
• W3161130147 hasConcept C108583219 @default.
• W3161130147 hasConcept C119857082 @default.
• W3161130147 hasConcept C127413603 @default.
• W3161130147 hasConcept C13280743 @default.
• W3161130147 hasConcept C154945302 @default.
• W3161130147 hasConcept C168065819 @default.
• W3161130147 hasConcept C17744445 @default.
• W3161130147 hasConcept C178790620 @default.
• W3161130147 hasConcept C185592680 @default.
• W3161130147 hasConcept C185798385 @default.
• W3161130147 hasConcept C199360897 @default.
• W3161130147 hasConcept C199539241 @default.
• W3161130147 hasConcept C200601418 @default.
• W3161130147 hasConcept C204495577 @default.
• W3161130147 hasConcept C205649164 @default.
• W3161130147 hasConcept C2776359362 @default.
• W3161130147 hasConcept C2777904410 @default.
• W3161130147 hasConcept C2779227376 @default.
• W3161130147 hasConcept C2984842247 @default.
• W3161130147 hasConcept C41008148 @default.
• W3161130147 hasConcept C50644808 @default.
• W3161130147 hasConcept C84945661 @default.
• W3161130147 hasConcept C94625758 @default.
• W3161130147 hasConcept C94966114 @default.
• W3161130147 hasConceptScore W3161130147C108583219 @default.
• W3161130147 hasConceptScore W3161130147C119857082 @default.
• W3161130147 hasConceptScore W3161130147C127413603 @default.
• W3161130147 hasConceptScore W3161130147C13280743 @default.
• W3161130147 hasConceptScore W3161130147C154945302 @default.
• W3161130147 hasConceptScore W3161130147C168065819 @default.
• W3161130147 hasConceptScore W3161130147C17744445 @default.
• W3161130147 hasConceptScore W3161130147C178790620 @default.
• W3161130147 hasConceptScore W3161130147C185592680 @default.
• W3161130147 hasConceptScore W3161130147C185798385 @default.
• W3161130147 hasConceptScore W3161130147C199360897 @default.
• W3161130147 hasConceptScore W3161130147C199539241 @default.
• W3161130147 hasConceptScore W3161130147C200601418 @default.
• W3161130147 hasConceptScore W3161130147C204495577 @default.
• W3161130147 hasConceptScore W3161130147C205649164 @default.
• W3161130147 hasConceptScore W3161130147C2776359362 @default.
• W3161130147 hasConceptScore W3161130147C2777904410 @default.
• W3161130147 hasConceptScore W3161130147C2779227376 @default.
• W3161130147 hasConceptScore W3161130147C2984842247 @default.
• W3161130147 hasConceptScore W3161130147C41008148 @default.
• W3161130147 hasConceptScore W3161130147C50644808 @default.
• W3161130147 hasConceptScore W3161130147C84945661 @default.
• W3161130147 hasConceptScore W3161130147C94625758 @default.
• W3161130147 hasConceptScore W3161130147C94966114 @default.
• W3161130147 hasLocation W31611301471 @default.
• W3161130147 hasOpenAccess W3161130147 @default.
• W3161130147 hasPrimaryLocation W31611301471 @default.
• W3161130147 hasRelatedWork W1973304445 @default.
• W3161130147 hasRelatedWork W2024088899 @default.
• W3161130147 hasRelatedWork W2058230131 @default.

• next