Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287726654> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4287726654 abstract "Deep neural networks are capable of training fast and generalizing well within many domains. Despite their promising performance, deep networks have shown sensitivities to perturbations of their inputs (e.g., adversarial examples) and their learned feature representations are often difficult to interpret, raising concerns about their true capability and trustworthiness. Recent work in adversarial training, a form of robust optimization in which the model is optimized against adversarial examples, demonstrates the ability to improve performance sensitivities to perturbations and yield feature representations that are more interpretable. Adversarial training, however, comes with an increased computational cost over that of standard (i.e., nonrobust) training, rendering it impractical for use in large-scale problems. Recent work suggests that a fast approximation to adversarial training shows promise for reducing training time and maintaining robustness in the presence of perturbations bounded by the infinity norm. In this work, we demonstrate that this approach extends to the Euclidean norm and preserves the human-aligned feature representations that are common for robust models. Additionally, we show that using a distributed training scheme can further reduce the time to train robust deep networks. Fast adversarial training is a promising approach that will provide increased security and explainability in machine learning applications for which robust optimization was previously thought to be impractical." @default.
- W4287726654 created "2022-07-26" @default.
- W4287726654 creator A5021196861 @default.
- W4287726654 creator A5063318273 @default.
- W4287726654 creator A5064725967 @default.
- W4287726654 date "2020-07-07" @default.
- W4287726654 modified "2023-10-17" @default.
- W4287726654 title "Fast Training of Deep Neural Networks Robust to Adversarial Perturbations" @default.
- W4287726654 doi "https://doi.org/10.48550/arxiv.2007.03832" @default.
- W4287726654 hasPublicationYear "2020" @default.
- W4287726654 type Work @default.
- W4287726654 citedByCount "0" @default.
- W4287726654 crossrefType "posted-content" @default.
- W4287726654 hasAuthorship W4287726654A5021196861 @default.
- W4287726654 hasAuthorship W4287726654A5063318273 @default.
- W4287726654 hasAuthorship W4287726654A5064725967 @default.
- W4287726654 hasBestOaLocation W42877266541 @default.
- W4287726654 hasConcept C104317684 @default.
- W4287726654 hasConcept C108583219 @default.
- W4287726654 hasConcept C119857082 @default.
- W4287726654 hasConcept C126255220 @default.
- W4287726654 hasConcept C134306372 @default.
- W4287726654 hasConcept C154945302 @default.
- W4287726654 hasConcept C17744445 @default.
- W4287726654 hasConcept C185592680 @default.
- W4287726654 hasConcept C191795146 @default.
- W4287726654 hasConcept C199539241 @default.
- W4287726654 hasConcept C2984842247 @default.
- W4287726654 hasConcept C33923547 @default.
- W4287726654 hasConcept C34388435 @default.
- W4287726654 hasConcept C37736160 @default.
- W4287726654 hasConcept C41008148 @default.
- W4287726654 hasConcept C50644808 @default.
- W4287726654 hasConcept C55493867 @default.
- W4287726654 hasConcept C63479239 @default.
- W4287726654 hasConceptScore W4287726654C104317684 @default.
- W4287726654 hasConceptScore W4287726654C108583219 @default.
- W4287726654 hasConceptScore W4287726654C119857082 @default.
- W4287726654 hasConceptScore W4287726654C126255220 @default.
- W4287726654 hasConceptScore W4287726654C134306372 @default.
- W4287726654 hasConceptScore W4287726654C154945302 @default.
- W4287726654 hasConceptScore W4287726654C17744445 @default.
- W4287726654 hasConceptScore W4287726654C185592680 @default.
- W4287726654 hasConceptScore W4287726654C191795146 @default.
- W4287726654 hasConceptScore W4287726654C199539241 @default.
- W4287726654 hasConceptScore W4287726654C2984842247 @default.
- W4287726654 hasConceptScore W4287726654C33923547 @default.
- W4287726654 hasConceptScore W4287726654C34388435 @default.
- W4287726654 hasConceptScore W4287726654C37736160 @default.
- W4287726654 hasConceptScore W4287726654C41008148 @default.
- W4287726654 hasConceptScore W4287726654C50644808 @default.
- W4287726654 hasConceptScore W4287726654C55493867 @default.
- W4287726654 hasConceptScore W4287726654C63479239 @default.
- W4287726654 hasLocation W42877266541 @default.
- W4287726654 hasOpenAccess W4287726654 @default.
- W4287726654 hasPrimaryLocation W42877266541 @default.
- W4287726654 hasRelatedWork W11339170 @default.
- W4287726654 hasRelatedWork W13678974 @default.
- W4287726654 hasRelatedWork W1407330 @default.
- W4287726654 hasRelatedWork W14627415 @default.
- W4287726654 hasRelatedWork W2203340 @default.
- W4287726654 hasRelatedWork W349990 @default.
- W4287726654 hasRelatedWork W8021486 @default.
- W4287726654 hasRelatedWork W8595301 @default.
- W4287726654 hasRelatedWork W9190101 @default.
- W4287726654 hasRelatedWork W933276 @default.
- W4287726654 isParatext "false" @default.
- W4287726654 isRetracted "false" @default.
- W4287726654 workType "article" @default.