Matches in SemOpenAlex for { <https://semopenalex.org/work/W4380714907> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W4380714907 abstract "Over the last decade, deep neural networks have achieved state of the art in computer vision tasks. These models, however, are susceptible to unusual inputs, known as adversarial examples, that cause them to misclassify or otherwise fail to detect objects. Here, we provide evidence that the increasing success of adversarial attacks is primarily due to increasing their size. We then demonstrate a method for generating the largest possible adversarial patch by building a adversarial pattern out of repeatable elements. This approach achieves a new state of the art in evading detection by YOLOv2 and YOLOv3. Finally, we present an experiment that fails to replicate the prior success of several attacks published in this field, and end with some comments on testing and reproducibility." @default.
- W4380714907 created "2023-06-15" @default.
- W4380714907 creator A5019828580 @default.
- W4380714907 date "2023-06-13" @default.
- W4380714907 modified "2023-09-23" @default.
- W4380714907 title "Area is all you need: repeatable elements make stronger adversarial attacks" @default.
- W4380714907 doi "https://doi.org/10.48550/arxiv.2306.07768" @default.
- W4380714907 hasPublicationYear "2023" @default.
- W4380714907 type Work @default.
- W4380714907 citedByCount "0" @default.
- W4380714907 crossrefType "posted-content" @default.
- W4380714907 hasAuthorship W4380714907A5019828580 @default.
- W4380714907 hasBestOaLocation W43807149071 @default.
- W4380714907 hasConcept C105795698 @default.
- W4380714907 hasConcept C108583219 @default.
- W4380714907 hasConcept C11413529 @default.
- W4380714907 hasConcept C119857082 @default.
- W4380714907 hasConcept C154945302 @default.
- W4380714907 hasConcept C202444582 @default.
- W4380714907 hasConcept C2522767166 @default.
- W4380714907 hasConcept C2781162219 @default.
- W4380714907 hasConcept C2984842247 @default.
- W4380714907 hasConcept C33923547 @default.
- W4380714907 hasConcept C37736160 @default.
- W4380714907 hasConcept C38652104 @default.
- W4380714907 hasConcept C41008148 @default.
- W4380714907 hasConcept C48103436 @default.
- W4380714907 hasConcept C9652623 @default.
- W4380714907 hasConceptScore W4380714907C105795698 @default.
- W4380714907 hasConceptScore W4380714907C108583219 @default.
- W4380714907 hasConceptScore W4380714907C11413529 @default.
- W4380714907 hasConceptScore W4380714907C119857082 @default.
- W4380714907 hasConceptScore W4380714907C154945302 @default.
- W4380714907 hasConceptScore W4380714907C202444582 @default.
- W4380714907 hasConceptScore W4380714907C2522767166 @default.
- W4380714907 hasConceptScore W4380714907C2781162219 @default.
- W4380714907 hasConceptScore W4380714907C2984842247 @default.
- W4380714907 hasConceptScore W4380714907C33923547 @default.
- W4380714907 hasConceptScore W4380714907C37736160 @default.
- W4380714907 hasConceptScore W4380714907C38652104 @default.
- W4380714907 hasConceptScore W4380714907C41008148 @default.
- W4380714907 hasConceptScore W4380714907C48103436 @default.
- W4380714907 hasConceptScore W4380714907C9652623 @default.
- W4380714907 hasLocation W43807149071 @default.
- W4380714907 hasOpenAccess W4380714907 @default.
- W4380714907 hasPrimaryLocation W43807149071 @default.
- W4380714907 hasRelatedWork W2791691546 @default.
- W4380714907 hasRelatedWork W2950066684 @default.
- W4380714907 hasRelatedWork W2952919291 @default.
- W4380714907 hasRelatedWork W3193857078 @default.
- W4380714907 hasRelatedWork W3208304128 @default.
- W4380714907 hasRelatedWork W4293054861 @default.
- W4380714907 hasRelatedWork W4298388782 @default.
- W4380714907 hasRelatedWork W4300837091 @default.
- W4380714907 hasRelatedWork W4312831135 @default.
- W4380714907 hasRelatedWork W4379255972 @default.
- W4380714907 isParatext "false" @default.
- W4380714907 isRetracted "false" @default.
- W4380714907 workType "article" @default.