Matches in SemOpenAlex for { <https://semopenalex.org/work/W4379984504> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4379984504 abstract "Convolutional Neural Networks (CNNs) have received great attention in the computer vision domain. However, CNNs are vulnerable to adversarial attacks, which are manipulations of input data that are imperceptible to humans but can fool the network. Several studies tried to address this issue, which can be divided into two categories: (i) training the network with adversarial examples, and (ii) optimizing the network architecture and/or hyperparameters. Although adversarial training is a sufficient defense mechanism, they suffer from requiring a large volume of training samples to cover a wide perturbation bound. Tweaking network activation functions (AFs) has been shown to provide promising results where CNNs suffer from performance loss. However, optimizing network AFs for compensating the negative impacts of adversarial attacks has not been addressed in the literature. This paper proposes the idea of searching for AFs that are robust against adversarial attacks. To this aim, we leverage the Simulated Annealing (SA) algorithm with a fast convergence time. This proposed method is called SARAF. We demonstrate the consistent effectiveness of SARAF by achieving up to 16.92%, 18.3%, and 15.57% accuracy improvement against BIM, FGSM, and PGD adversarial attacks, respectively, over ResNet-18 with ReLU AFs (baseline) trained on CIFAR-10. Meanwhile, SARAF provides a significant search efficiency compared to random search as the optimization baseline." @default.
- W4379984504 created "2023-06-10" @default.
- W4379984504 creator A5001642408 @default.
- W4379984504 creator A5008146469 @default.
- W4379984504 creator A5044187839 @default.
- W4379984504 creator A5078919695 @default.
- W4379984504 creator A5089443492 @default.
- W4379984504 date "2023-03-10" @default.
- W4379984504 modified "2023-09-24" @default.
- W4379984504 title "SARAF: Searching for Adversarial Robust Activation Functions" @default.
- W4379984504 cites W1977207509 @default.
- W4379984504 cites W2194775991 @default.
- W4379984504 cites W2201305792 @default.
- W4379984504 cites W2243397390 @default.
- W4379984504 cites W2293768274 @default.
- W4379984504 cites W2963626858 @default.
- W4379984504 cites W2964081807 @default.
- W4379984504 cites W2967706363 @default.
- W4379984504 cites W2970585064 @default.
- W4379984504 cites W2972401431 @default.
- W4379984504 cites W2981954928 @default.
- W4379984504 cites W2999552082 @default.
- W4379984504 cites W3035239274 @default.
- W4379984504 cites W3038944566 @default.
- W4379984504 cites W3083251833 @default.
- W4379984504 cites W3135034047 @default.
- W4379984504 cites W3163963286 @default.
- W4379984504 cites W3205297166 @default.
- W4379984504 cites W3212981375 @default.
- W4379984504 cites W4283697320 @default.
- W4379984504 cites W4312664201 @default.
- W4379984504 doi "https://doi.org/10.1145/3589572.3589598" @default.
- W4379984504 hasPublicationYear "2023" @default.
- W4379984504 type Work @default.
- W4379984504 citedByCount "0" @default.
- W4379984504 crossrefType "proceedings-article" @default.
- W4379984504 hasAuthorship W4379984504A5001642408 @default.
- W4379984504 hasAuthorship W4379984504A5008146469 @default.
- W4379984504 hasAuthorship W4379984504A5044187839 @default.
- W4379984504 hasAuthorship W4379984504A5078919695 @default.
- W4379984504 hasAuthorship W4379984504A5089443492 @default.
- W4379984504 hasBestOaLocation W43799845041 @default.
- W4379984504 hasConcept C111368507 @default.
- W4379984504 hasConcept C111919701 @default.
- W4379984504 hasConcept C119857082 @default.
- W4379984504 hasConcept C12725497 @default.
- W4379984504 hasConcept C127313418 @default.
- W4379984504 hasConcept C153083717 @default.
- W4379984504 hasConcept C154945302 @default.
- W4379984504 hasConcept C2780200862 @default.
- W4379984504 hasConcept C37736160 @default.
- W4379984504 hasConcept C41008148 @default.
- W4379984504 hasConcept C81363708 @default.
- W4379984504 hasConcept C8642999 @default.
- W4379984504 hasConceptScore W4379984504C111368507 @default.
- W4379984504 hasConceptScore W4379984504C111919701 @default.
- W4379984504 hasConceptScore W4379984504C119857082 @default.
- W4379984504 hasConceptScore W4379984504C12725497 @default.
- W4379984504 hasConceptScore W4379984504C127313418 @default.
- W4379984504 hasConceptScore W4379984504C153083717 @default.
- W4379984504 hasConceptScore W4379984504C154945302 @default.
- W4379984504 hasConceptScore W4379984504C2780200862 @default.
- W4379984504 hasConceptScore W4379984504C37736160 @default.
- W4379984504 hasConceptScore W4379984504C41008148 @default.
- W4379984504 hasConceptScore W4379984504C81363708 @default.
- W4379984504 hasConceptScore W4379984504C8642999 @default.
- W4379984504 hasLocation W43799845041 @default.
- W4379984504 hasOpenAccess W4379984504 @default.
- W4379984504 hasPrimaryLocation W43799845041 @default.
- W4379984504 hasRelatedWork W2784294038 @default.
- W4379984504 hasRelatedWork W3027997911 @default.
- W4379984504 hasRelatedWork W3175409811 @default.
- W4379984504 hasRelatedWork W4210794429 @default.
- W4379984504 hasRelatedWork W4280535922 @default.
- W4379984504 hasRelatedWork W4287776258 @default.
- W4379984504 hasRelatedWork W4295309597 @default.
- W4379984504 hasRelatedWork W4309113015 @default.
- W4379984504 hasRelatedWork W4312411443 @default.
- W4379984504 hasRelatedWork W4379255972 @default.
- W4379984504 isParatext "false" @default.
- W4379984504 isRetracted "false" @default.
- W4379984504 workType "article" @default.