Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287690173> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4287690173 abstract "Sensitivity to adversarial noise hinders deployment of machine learning algorithms in security-critical applications. Although many adversarial defenses have been proposed, robustness to adversarial noise remains an open problem. The most compelling defense, adversarial training, requires a substantial increase in processing time and it has been shown to overfit on the training data. In this paper, we aim to overcome these limitations by training robust models in low data regimes and transfer adversarial knowledge between different models. We train a meta-optimizer which learns to robustly optimize a model using adversarial examples and is able to transfer the knowledge learned to new models, without the need to generate new adversarial examples. Experimental results show the meta-optimizer is consistent across different architectures and data sets, suggesting it is possible to automatically patch adversarial vulnerabilities." @default.
- W4287690173 created "2022-07-26" @default.
- W4287690173 creator A5018169511 @default.
- W4287690173 creator A5027550247 @default.
- W4287690173 creator A5049830358 @default.
- W4287690173 date "2020-08-12" @default.
- W4287690173 modified "2023-09-24" @default.
- W4287690173 title "Learning to Learn from Mistakes: Robust Optimization for Adversarial Noise" @default.
- W4287690173 doi "https://doi.org/10.48550/arxiv.2008.05247" @default.
- W4287690173 hasPublicationYear "2020" @default.
- W4287690173 type Work @default.
- W4287690173 citedByCount "0" @default.
- W4287690173 crossrefType "posted-content" @default.
- W4287690173 hasAuthorship W4287690173A5018169511 @default.
- W4287690173 hasAuthorship W4287690173A5027550247 @default.
- W4287690173 hasAuthorship W4287690173A5049830358 @default.
- W4287690173 hasBestOaLocation W42876901731 @default.
- W4287690173 hasConcept C104317684 @default.
- W4287690173 hasConcept C105339364 @default.
- W4287690173 hasConcept C111919701 @default.
- W4287690173 hasConcept C115961682 @default.
- W4287690173 hasConcept C119857082 @default.
- W4287690173 hasConcept C154945302 @default.
- W4287690173 hasConcept C185592680 @default.
- W4287690173 hasConcept C22019652 @default.
- W4287690173 hasConcept C2778403875 @default.
- W4287690173 hasConcept C37736160 @default.
- W4287690173 hasConcept C41008148 @default.
- W4287690173 hasConcept C50644808 @default.
- W4287690173 hasConcept C55493867 @default.
- W4287690173 hasConcept C63479239 @default.
- W4287690173 hasConcept C99498987 @default.
- W4287690173 hasConceptScore W4287690173C104317684 @default.
- W4287690173 hasConceptScore W4287690173C105339364 @default.
- W4287690173 hasConceptScore W4287690173C111919701 @default.
- W4287690173 hasConceptScore W4287690173C115961682 @default.
- W4287690173 hasConceptScore W4287690173C119857082 @default.
- W4287690173 hasConceptScore W4287690173C154945302 @default.
- W4287690173 hasConceptScore W4287690173C185592680 @default.
- W4287690173 hasConceptScore W4287690173C22019652 @default.
- W4287690173 hasConceptScore W4287690173C2778403875 @default.
- W4287690173 hasConceptScore W4287690173C37736160 @default.
- W4287690173 hasConceptScore W4287690173C41008148 @default.
- W4287690173 hasConceptScore W4287690173C50644808 @default.
- W4287690173 hasConceptScore W4287690173C55493867 @default.
- W4287690173 hasConceptScore W4287690173C63479239 @default.
- W4287690173 hasConceptScore W4287690173C99498987 @default.
- W4287690173 hasLocation W42876901731 @default.
- W4287690173 hasOpenAccess W4287690173 @default.
- W4287690173 hasPrimaryLocation W42876901731 @default.
- W4287690173 hasRelatedWork W2610321374 @default.
- W4287690173 hasRelatedWork W2952919291 @default.
- W4287690173 hasRelatedWork W2963207607 @default.
- W4287690173 hasRelatedWork W2989932438 @default.
- W4287690173 hasRelatedWork W3034176567 @default.
- W4287690173 hasRelatedWork W3099765033 @default.
- W4287690173 hasRelatedWork W3204747554 @default.
- W4287690173 hasRelatedWork W4225586443 @default.
- W4287690173 hasRelatedWork W4310513940 @default.
- W4287690173 hasRelatedWork W4312741812 @default.
- W4287690173 isParatext "false" @default.
- W4287690173 isRetracted "false" @default.
- W4287690173 workType "article" @default.