FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3093484254> ?p ?o ?g. }

• W3093484254 abstract "Deep neural networks have achieved state-of-the-art performance in a variety of fields. Recent works observe that a class of widely used neural networks can be viewed as the Euler method of numerical discretization. From the numerical discretization perspective, Strong Stability Preserving (SSP) methods are more advanced techniques than the explicit Euler method that produce both accurate and stable solutions. Motivated by the SSP property and a generalized Runge-Kutta method, we propose Strong Stability Preserving networks (SSP networks) which improve robustness against adversarial attacks. We empirically demonstrate that the proposed networks improve the robustness against adversarial examples without any defensive methods. Further, the SSP networks are complementary with a state-of-the-art adversarial training scheme. Lastly, our experiments show that SSP networks suppress the blow-up of adversarial perturbations. Our results open up a way to study robust architectures of neural networks leveraging rich knowledge from numerical discretization literature." @default.
• W3093484254 created "2020-10-29" @default.
• W3093484254 creator A5043731528 @default.
• W3093484254 creator A5066319071 @default.
• W3093484254 creator A5068799024 @default.
• W3093484254 creator A5076917278 @default.
• W3093484254 creator A5084814930 @default.
• W3093484254 creator A5085177012 @default.
• W3093484254 date "2020-10-20" @default.
• W3093484254 modified "2023-09-23" @default.
• W3093484254 title "Robust Neural Networks inspired by Strong Stability Preserving Runge-Kutta methods" @default.
• W3093484254 cites W1673923490 @default.
• W3093484254 cites W1677182931 @default.
• W3093484254 cites W1969647398 @default.
• W3093484254 cites W1998830482 @default.
• W3093484254 cites W2054662916 @default.
• W3093484254 cites W2093711514 @default.
• W3093484254 cites W2116729868 @default.
• W3093484254 cites W2153417333 @default.
• W3093484254 cites W2194775991 @default.
• W3093484254 cites W2302255633 @default.
• W3093484254 cites W2786118190 @default.
• W3093484254 cites W2895434480 @default.
• W3093484254 cites W2945793108 @default.
• W3093484254 cites W2951165433 @default.
• W3093484254 cites W2954978443 @default.
• W3093484254 cites W2958796454 @default.
• W3093484254 cites W2962729158 @default.
• W3093484254 cites W2962737770 @default.
• W3093484254 cites W2962759300 @default.
• W3093484254 cites W2962872506 @default.
• W3093484254 cites W2962943487 @default.
• W3093484254 cites W2963004026 @default.
• W3093484254 cites W2963143631 @default.
• W3093484254 cites W2963207607 @default.
• W3093484254 cites W2963302407 @default.
• W3093484254 cites W2963496101 @default.
• W3093484254 cites W2963755523 @default.
• W3093484254 cites W2963857521 @default.
• W3093484254 cites W2964082701 @default.
• W3093484254 cites W2964116600 @default.
• W3093484254 cites W2964121744 @default.
• W3093484254 cites W2964197269 @default.
• W3093484254 cites W2964253222 @default.
• W3093484254 cites W2974916071 @default.
• W3093484254 cites W3118608800 @default.
• W3093484254 doi "https://doi.org/10.48550/arxiv.2010.10047" @default.
• W3093484254 hasPublicationYear "2020" @default.
• W3093484254 type Work @default.
• W3093484254 sameAs 3093484254 @default.
• W3093484254 citedByCount "0" @default.
• W3093484254 crossrefType "posted-content" @default.
• W3093484254 hasAuthorship W3093484254A5043731528 @default.
• W3093484254 hasAuthorship W3093484254A5066319071 @default.
• W3093484254 hasAuthorship W3093484254A5068799024 @default.
• W3093484254 hasAuthorship W3093484254A5076917278 @default.
• W3093484254 hasAuthorship W3093484254A5084814930 @default.
• W3093484254 hasAuthorship W3093484254A5085177012 @default.
• W3093484254 hasBestOaLocation W30934842541 @default.
• W3093484254 hasConcept C104317684 @default.
• W3093484254 hasConcept C112972136 @default.
• W3093484254 hasConcept C119857082 @default.
• W3093484254 hasConcept C134306372 @default.
• W3093484254 hasConcept C154945302 @default.
• W3093484254 hasConcept C181582579 @default.
• W3093484254 hasConcept C185592680 @default.
• W3093484254 hasConcept C2984842247 @default.
• W3093484254 hasConcept C33923547 @default.
• W3093484254 hasConcept C37736160 @default.
• W3093484254 hasConcept C41008148 @default.
• W3093484254 hasConcept C48753275 @default.
• W3093484254 hasConcept C50644808 @default.
• W3093484254 hasConcept C55493867 @default.
• W3093484254 hasConcept C62884695 @default.
• W3093484254 hasConcept C63479239 @default.
• W3093484254 hasConcept C73000952 @default.
• W3093484254 hasConcept C75380026 @default.
• W3093484254 hasConceptScore W3093484254C104317684 @default.
• W3093484254 hasConceptScore W3093484254C112972136 @default.
• W3093484254 hasConceptScore W3093484254C119857082 @default.
• W3093484254 hasConceptScore W3093484254C134306372 @default.
• W3093484254 hasConceptScore W3093484254C154945302 @default.
• W3093484254 hasConceptScore W3093484254C181582579 @default.
• W3093484254 hasConceptScore W3093484254C185592680 @default.
• W3093484254 hasConceptScore W3093484254C2984842247 @default.
• W3093484254 hasConceptScore W3093484254C33923547 @default.
• W3093484254 hasConceptScore W3093484254C37736160 @default.
• W3093484254 hasConceptScore W3093484254C41008148 @default.
• W3093484254 hasConceptScore W3093484254C48753275 @default.
• W3093484254 hasConceptScore W3093484254C50644808 @default.
• W3093484254 hasConceptScore W3093484254C55493867 @default.
• W3093484254 hasConceptScore W3093484254C62884695 @default.
• W3093484254 hasConceptScore W3093484254C63479239 @default.
• W3093484254 hasConceptScore W3093484254C73000952 @default.
• W3093484254 hasConceptScore W3093484254C75380026 @default.
• W3093484254 hasLocation W30934842541 @default.
• W3093484254 hasOpenAccess W3093484254 @default.
• W3093484254 hasPrimaryLocation W30934842541 @default.
• W3093484254 hasRelatedWork W2045654799 @default.
• W3093484254 hasRelatedWork W2112489823 @default.

• next