Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950942529> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W2950942529 abstract "Adversarial samples are perturbed inputs crafted to mislead the machine learning systems. A training mechanism, called adversarial training, which presents adversarial samples along with clean samples has been introduced to learn robust models. In order to scale adversarial training for large datasets, these perturbations can only be crafted using fast and simple methods (e.g., gradient ascent). However, it is shown that adversarial training converges to a degenerate minimum, where the model appears to be robust by generating weaker adversaries. As a result, the models are vulnerable to simple black-box attacks. In this paper we, (i) demonstrate the shortcomings of existing evaluation policy, (ii) introduce novel variants of white-box and black-box attacks, dubbed gray-box adversarial attacks based on which we propose novel evaluation method to assess the robustness of the learned models, and (iii) propose a novel variant of adversarial training, named Graybox Adversarial Training that uses intermediate versions of the models to seed the adversaries. Experimental evaluation demonstrates that the models trained using our method exhibit better robustness compared to both undefended and adversarially trained model" @default.
- W2950942529 created "2019-06-27" @default.
- W2950942529 creator A5002522068 @default.
- W2950942529 creator A5024899038 @default.
- W2950942529 creator A5062096014 @default.
- W2950942529 date "2018-08-06" @default.
- W2950942529 modified "2023-09-25" @default.
- W2950942529 title "Gray-box Adversarial Training" @default.
- W2950942529 cites W2095577883 @default.
- W2950942529 cites W2963207607 @default.
- W2950942529 cites W9657784 @default.
- W2950942529 hasPublicationYear "2018" @default.
- W2950942529 type Work @default.
- W2950942529 sameAs 2950942529 @default.
- W2950942529 citedByCount "1" @default.
- W2950942529 countsByYear W29509425292021 @default.
- W2950942529 crossrefType "posted-content" @default.
- W2950942529 hasAuthorship W2950942529A5002522068 @default.
- W2950942529 hasAuthorship W2950942529A5024899038 @default.
- W2950942529 hasAuthorship W2950942529A5062096014 @default.
- W2950942529 hasConcept C104317684 @default.
- W2950942529 hasConcept C119857082 @default.
- W2950942529 hasConcept C121332964 @default.
- W2950942529 hasConcept C153294291 @default.
- W2950942529 hasConcept C154945302 @default.
- W2950942529 hasConcept C180932941 @default.
- W2950942529 hasConcept C185592680 @default.
- W2950942529 hasConcept C2777211547 @default.
- W2950942529 hasConcept C37736160 @default.
- W2950942529 hasConcept C41008148 @default.
- W2950942529 hasConcept C51632099 @default.
- W2950942529 hasConcept C55493867 @default.
- W2950942529 hasConcept C62520636 @default.
- W2950942529 hasConcept C63479239 @default.
- W2950942529 hasConcept C72319582 @default.
- W2950942529 hasConcept C94966114 @default.
- W2950942529 hasConceptScore W2950942529C104317684 @default.
- W2950942529 hasConceptScore W2950942529C119857082 @default.
- W2950942529 hasConceptScore W2950942529C121332964 @default.
- W2950942529 hasConceptScore W2950942529C153294291 @default.
- W2950942529 hasConceptScore W2950942529C154945302 @default.
- W2950942529 hasConceptScore W2950942529C180932941 @default.
- W2950942529 hasConceptScore W2950942529C185592680 @default.
- W2950942529 hasConceptScore W2950942529C2777211547 @default.
- W2950942529 hasConceptScore W2950942529C37736160 @default.
- W2950942529 hasConceptScore W2950942529C41008148 @default.
- W2950942529 hasConceptScore W2950942529C51632099 @default.
- W2950942529 hasConceptScore W2950942529C55493867 @default.
- W2950942529 hasConceptScore W2950942529C62520636 @default.
- W2950942529 hasConceptScore W2950942529C63479239 @default.
- W2950942529 hasConceptScore W2950942529C72319582 @default.
- W2950942529 hasConceptScore W2950942529C94966114 @default.
- W2950942529 hasLocation W29509425291 @default.
- W2950942529 hasOpenAccess W2950942529 @default.
- W2950942529 hasPrimaryLocation W29509425291 @default.
- W2950942529 hasRelatedWork W2777353073 @default.
- W2950942529 hasRelatedWork W2781758978 @default.
- W2950942529 hasRelatedWork W2804545960 @default.
- W2950942529 hasRelatedWork W2890591829 @default.
- W2950942529 hasRelatedWork W2963612069 @default.
- W2950942529 hasRelatedWork W2963870211 @default.
- W2950942529 hasRelatedWork W2995645057 @default.
- W2950942529 hasRelatedWork W3007712033 @default.
- W2950942529 hasRelatedWork W3018911559 @default.
- W2950942529 hasRelatedWork W3024886890 @default.
- W2950942529 hasRelatedWork W3028525609 @default.
- W2950942529 hasRelatedWork W3036303027 @default.
- W2950942529 hasRelatedWork W3039130562 @default.
- W2950942529 hasRelatedWork W3047745574 @default.
- W2950942529 hasRelatedWork W3093152201 @default.
- W2950942529 hasRelatedWork W3104032928 @default.
- W2950942529 hasRelatedWork W3122142687 @default.
- W2950942529 hasRelatedWork W3130773951 @default.
- W2950942529 hasRelatedWork W3198031492 @default.
- W2950942529 hasRelatedWork W3210431426 @default.
- W2950942529 isParatext "false" @default.
- W2950942529 isRetracted "false" @default.
- W2950942529 magId "2950942529" @default.
- W2950942529 workType "article" @default.