Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387294372> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4387294372 abstract "Artificial Neural Networks (ANNs) trained with Backpropagation (BP) show astounding performance and are increasingly often used in performing our daily life tasks. However, ANNs are highly vulnerable to adversarial attacks, which alter inputs with small targeted perturbations that drastically disrupt the models' performance. The most effective method to make ANNs robust against these attacks is adversarial training, in which the training dataset is augmented with exemplary adversarial samples. Unfortunately, this approach has the drawback of increased training complexity since generating adversarial samples is very computationally demanding. In contrast to ANNs, humans are not susceptible to adversarial attacks. Therefore, in this work, we investigate whether biologically-plausible learning algorithms are more robust against adversarial attacks than BP. In particular, we present an extensive comparative analysis of the adversarial robustness of BP and Present the Error to Perturb the Input To modulate Activity (PEPITA), a recently proposed biologically-plausible learning algorithm, on various computer vision tasks. We observe that PEPITA has higher intrinsic adversarial robustness and, with adversarial training, has a more favourable natural-vs-adversarial performance trade-off as, for the same natural accuracies, PEPITA's adversarial accuracies decrease in average by 0.26% and BP's by 8.05%." @default.
- W4387294372 created "2023-10-03" @default.
- W4387294372 creator A5001186358 @default.
- W4387294372 creator A5024779827 @default.
- W4387294372 creator A5025253322 @default.
- W4387294372 creator A5066680992 @default.
- W4387294372 creator A5090542476 @default.
- W4387294372 date "2023-09-29" @default.
- W4387294372 modified "2023-10-09" @default.
- W4387294372 title "Efficient Biologically Plausible Adversarial Training" @default.
- W4387294372 doi "https://doi.org/10.48550/arxiv.2309.17348" @default.
- W4387294372 hasPublicationYear "2023" @default.
- W4387294372 type Work @default.
- W4387294372 citedByCount "0" @default.
- W4387294372 crossrefType "posted-content" @default.
- W4387294372 hasAuthorship W4387294372A5001186358 @default.
- W4387294372 hasAuthorship W4387294372A5024779827 @default.
- W4387294372 hasAuthorship W4387294372A5025253322 @default.
- W4387294372 hasAuthorship W4387294372A5066680992 @default.
- W4387294372 hasAuthorship W4387294372A5090542476 @default.
- W4387294372 hasBestOaLocation W43872943721 @default.
- W4387294372 hasConcept C104317684 @default.
- W4387294372 hasConcept C119857082 @default.
- W4387294372 hasConcept C121332964 @default.
- W4387294372 hasConcept C153294291 @default.
- W4387294372 hasConcept C154945302 @default.
- W4387294372 hasConcept C155032097 @default.
- W4387294372 hasConcept C185592680 @default.
- W4387294372 hasConcept C2777211547 @default.
- W4387294372 hasConcept C2984842247 @default.
- W4387294372 hasConcept C37736160 @default.
- W4387294372 hasConcept C41008148 @default.
- W4387294372 hasConcept C50644808 @default.
- W4387294372 hasConcept C51632099 @default.
- W4387294372 hasConcept C55493867 @default.
- W4387294372 hasConcept C63479239 @default.
- W4387294372 hasConceptScore W4387294372C104317684 @default.
- W4387294372 hasConceptScore W4387294372C119857082 @default.
- W4387294372 hasConceptScore W4387294372C121332964 @default.
- W4387294372 hasConceptScore W4387294372C153294291 @default.
- W4387294372 hasConceptScore W4387294372C154945302 @default.
- W4387294372 hasConceptScore W4387294372C155032097 @default.
- W4387294372 hasConceptScore W4387294372C185592680 @default.
- W4387294372 hasConceptScore W4387294372C2777211547 @default.
- W4387294372 hasConceptScore W4387294372C2984842247 @default.
- W4387294372 hasConceptScore W4387294372C37736160 @default.
- W4387294372 hasConceptScore W4387294372C41008148 @default.
- W4387294372 hasConceptScore W4387294372C50644808 @default.
- W4387294372 hasConceptScore W4387294372C51632099 @default.
- W4387294372 hasConceptScore W4387294372C55493867 @default.
- W4387294372 hasConceptScore W4387294372C63479239 @default.
- W4387294372 hasLocation W43872943721 @default.
- W4387294372 hasOpenAccess W4387294372 @default.
- W4387294372 hasPrimaryLocation W43872943721 @default.
- W4387294372 hasRelatedWork W2738001131 @default.
- W4387294372 hasRelatedWork W2950183588 @default.
- W4387294372 hasRelatedWork W2997056298 @default.
- W4387294372 hasRelatedWork W3080754722 @default.
- W4387294372 hasRelatedWork W3093978547 @default.
- W4387294372 hasRelatedWork W3127875750 @default.
- W4387294372 hasRelatedWork W3203790781 @default.
- W4387294372 hasRelatedWork W4285785480 @default.
- W4387294372 hasRelatedWork W4383221314 @default.
- W4387294372 hasRelatedWork W4386850404 @default.
- W4387294372 isParatext "false" @default.
- W4387294372 isRetracted "false" @default.
- W4387294372 workType "article" @default.