Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367397356> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4367397356 endingPage "110370" @default.
- W4367397356 startingPage "110370" @default.
- W4367397356 abstract "Due to the importance of security, the adversarial attack has become an increasingly popular area for deep learning, especially the black-box adversarial attack, which can only obtain the input and output of the target model, resulting in closer to the real world. Query-based methods are a common strategy for performing black-box attacks, and they can find a slight perturbation to make target model misclassify by continuously querying the target model to obtain the output of the target model. However, query-based methods usually suffer from a serious flaw that needs massive queries, which is unaccepted in the real world. Although some query-efficient methods have been proposed to alleviate the above problems, they greatly sacrifice the quality of adversarial examples due to their problem formulations. To generate high-quality adversarial examples with a limited query budget, we propose a Bayesian evolutionary optimization (BEO) based black-box attack method using differential evolution, where a Gaussian processes model is employed to approximate the real objective function. As a key component of the BEO, we use seven acquisition functions to sample the new solution to update the Gaussian processes model, and an information entropy based selection strategy is proposed to adaptively choose the acquisition function. Finally, an effectiveness validation study is carried out comparing the proposed method with five other black-box attack methods and one Bayesian optimization (BO) method using the CIFAR-10 and ImageNet datasets. Experimental results demonstrate that the proposed method can effectively generate adversarial examples using only 200 queries." @default.
- W4367397356 created "2023-04-30" @default.
- W4367397356 creator A5003753200 @default.
- W4367397356 creator A5058451623 @default.
- W4367397356 creator A5068914676 @default.
- W4367397356 creator A5089866078 @default.
- W4367397356 creator A5091324372 @default.
- W4367397356 date "2023-07-01" @default.
- W4367397356 modified "2023-10-17" @default.
- W4367397356 title "Bayesian evolutionary optimization for crafting high-quality adversarial examples with limited query budget" @default.
- W4367397356 cites W1995875735 @default.
- W4367397356 cites W2011174137 @default.
- W4367397356 cites W2092211810 @default.
- W4367397356 cites W2136918060 @default.
- W4367397356 cites W2156194072 @default.
- W4367397356 cites W2194775991 @default.
- W4367397356 cites W2243397390 @default.
- W4367397356 cites W2603766943 @default.
- W4367397356 cites W2746600820 @default.
- W4367397356 cites W2758271972 @default.
- W4367397356 cites W2774644650 @default.
- W4367397356 cites W2798302089 @default.
- W4367397356 cites W2891140334 @default.
- W4367397356 cites W2962711307 @default.
- W4367397356 cites W2962847335 @default.
- W4367397356 cites W2963726920 @default.
- W4367397356 cites W2980361376 @default.
- W4367397356 cites W2999606423 @default.
- W4367397356 cites W3018911559 @default.
- W4367397356 cites W3046870880 @default.
- W4367397356 cites W3086146226 @default.
- W4367397356 cites W3103557498 @default.
- W4367397356 cites W3133211661 @default.
- W4367397356 cites W3153328918 @default.
- W4367397356 cites W3171288285 @default.
- W4367397356 cites W3192524834 @default.
- W4367397356 cites W3195153428 @default.
- W4367397356 cites W3201305891 @default.
- W4367397356 cites W4249517230 @default.
- W4367397356 doi "https://doi.org/10.1016/j.asoc.2023.110370" @default.
- W4367397356 hasPublicationYear "2023" @default.
- W4367397356 type Work @default.
- W4367397356 citedByCount "1" @default.
- W4367397356 crossrefType "journal-article" @default.
- W4367397356 hasAuthorship W4367397356A5003753200 @default.
- W4367397356 hasAuthorship W4367397356A5058451623 @default.
- W4367397356 hasAuthorship W4367397356A5068914676 @default.
- W4367397356 hasAuthorship W4367397356A5089866078 @default.
- W4367397356 hasAuthorship W4367397356A5091324372 @default.
- W4367397356 hasConcept C107673813 @default.
- W4367397356 hasConcept C119857082 @default.
- W4367397356 hasConcept C121332964 @default.
- W4367397356 hasConcept C124101348 @default.
- W4367397356 hasConcept C154945302 @default.
- W4367397356 hasConcept C163716315 @default.
- W4367397356 hasConcept C2778049539 @default.
- W4367397356 hasConcept C37736160 @default.
- W4367397356 hasConcept C41008148 @default.
- W4367397356 hasConcept C61326573 @default.
- W4367397356 hasConcept C62520636 @default.
- W4367397356 hasConcept C74750220 @default.
- W4367397356 hasConcept C94966114 @default.
- W4367397356 hasConceptScore W4367397356C107673813 @default.
- W4367397356 hasConceptScore W4367397356C119857082 @default.
- W4367397356 hasConceptScore W4367397356C121332964 @default.
- W4367397356 hasConceptScore W4367397356C124101348 @default.
- W4367397356 hasConceptScore W4367397356C154945302 @default.
- W4367397356 hasConceptScore W4367397356C163716315 @default.
- W4367397356 hasConceptScore W4367397356C2778049539 @default.
- W4367397356 hasConceptScore W4367397356C37736160 @default.
- W4367397356 hasConceptScore W4367397356C41008148 @default.
- W4367397356 hasConceptScore W4367397356C61326573 @default.
- W4367397356 hasConceptScore W4367397356C62520636 @default.
- W4367397356 hasConceptScore W4367397356C74750220 @default.
- W4367397356 hasConceptScore W4367397356C94966114 @default.
- W4367397356 hasFunder F4320321001 @default.
- W4367397356 hasLocation W43673973561 @default.
- W4367397356 hasOpenAccess W4367397356 @default.
- W4367397356 hasPrimaryLocation W43673973561 @default.
- W4367397356 hasRelatedWork W2724199550 @default.
- W4367397356 hasRelatedWork W2936660777 @default.
- W4367397356 hasRelatedWork W2977187670 @default.
- W4367397356 hasRelatedWork W3042656994 @default.
- W4367397356 hasRelatedWork W3155731460 @default.
- W4367397356 hasRelatedWork W3199608561 @default.
- W4367397356 hasRelatedWork W4280525836 @default.
- W4367397356 hasRelatedWork W4287723950 @default.
- W4367397356 hasRelatedWork W4384929195 @default.
- W4367397356 hasRelatedWork W4384932390 @default.
- W4367397356 hasVolume "142" @default.
- W4367397356 isParatext "false" @default.
- W4367397356 isRetracted "false" @default.
- W4367397356 workType "article" @default.