Matches in SemOpenAlex for { <https://semopenalex.org/work/W2909607453> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2909607453 abstract "Deep learning has been achieving top performance in many tasks. Since training of a deep learning model requires a great deal of cost, we need to treat neural network models as valuable intellectual properties. One concern in such a situation is that some malicious user might redistribute the model or provide a prediction service using the model without permission. One promising solution is digital watermarking, to embed a mechanism into the model so that the owner of the model can verify the ownership of the model externally. In this study, we present a novel attack method against watermark, query modification, and demonstrate that all of the existing watermark methods are vulnerable to either of query modification or existing attack method (model modification). To overcome this vulnerability, we present a novel watermarking method, exponential weighting. We experimentally show that our watermarking method achieves high verification performance of watermark even under a malicious attempt of unauthorized service providers, such as model modification and query modification, without sacrificing the predictive performance of the neural network model." @default.
- W2909607453 created "2019-01-25" @default.
- W2909607453 creator A5022139141 @default.
- W2909607453 creator A5083024814 @default.
- W2909607453 date "2019-01-18" @default.
- W2909607453 modified "2023-09-27" @default.
- W2909607453 title "Robust Watermarking of Neural Network with Exponential Weighting" @default.
- W2909607453 cites W1686810756 @default.
- W2909607453 cites W1922655562 @default.
- W2909607453 cites W2002427601 @default.
- W2909607453 cites W2100495367 @default.
- W2909607453 cites W2149933564 @default.
- W2909607453 cites W2193413348 @default.
- W2909607453 cites W2194775991 @default.
- W2909607453 cites W2579318729 @default.
- W2909607453 cites W2590796488 @default.
- W2909607453 cites W2768064608 @default.
- W2909607453 cites W2796299376 @default.
- W2909607453 cites W2806082141 @default.
- W2909607453 cites W2899585792 @default.
- W2909607453 cites W2963207607 @default.
- W2909607453 cites W2963674932 @default.
- W2909607453 cites W2963771448 @default.
- W2909607453 cites W2964153729 @default.
- W2909607453 cites W2964159205 @default.
- W2909607453 cites W2964308564 @default.
- W2909607453 cites W3118608800 @default.
- W2909607453 doi "https://doi.org/10.48550/arxiv.1901.06151" @default.
- W2909607453 hasPublicationYear "2019" @default.
- W2909607453 type Work @default.
- W2909607453 sameAs 2909607453 @default.
- W2909607453 citedByCount "6" @default.
- W2909607453 countsByYear W29096074532019 @default.
- W2909607453 countsByYear W29096074532020 @default.
- W2909607453 countsByYear W29096074532021 @default.
- W2909607453 crossrefType "posted-content" @default.
- W2909607453 hasAuthorship W2909607453A5022139141 @default.
- W2909607453 hasAuthorship W2909607453A5083024814 @default.
- W2909607453 hasBestOaLocation W29096074531 @default.
- W2909607453 hasConcept C108583219 @default.
- W2909607453 hasConcept C115961682 @default.
- W2909607453 hasConcept C119857082 @default.
- W2909607453 hasConcept C124101348 @default.
- W2909607453 hasConcept C126838900 @default.
- W2909607453 hasConcept C150817343 @default.
- W2909607453 hasConcept C154945302 @default.
- W2909607453 hasConcept C164112704 @default.
- W2909607453 hasConcept C17744445 @default.
- W2909607453 hasConcept C183115368 @default.
- W2909607453 hasConcept C199539241 @default.
- W2909607453 hasConcept C2779089604 @default.
- W2909607453 hasConcept C38652104 @default.
- W2909607453 hasConcept C41008148 @default.
- W2909607453 hasConcept C41608201 @default.
- W2909607453 hasConcept C50644808 @default.
- W2909607453 hasConcept C65856478 @default.
- W2909607453 hasConcept C71924100 @default.
- W2909607453 hasConcept C95713431 @default.
- W2909607453 hasConceptScore W2909607453C108583219 @default.
- W2909607453 hasConceptScore W2909607453C115961682 @default.
- W2909607453 hasConceptScore W2909607453C119857082 @default.
- W2909607453 hasConceptScore W2909607453C124101348 @default.
- W2909607453 hasConceptScore W2909607453C126838900 @default.
- W2909607453 hasConceptScore W2909607453C150817343 @default.
- W2909607453 hasConceptScore W2909607453C154945302 @default.
- W2909607453 hasConceptScore W2909607453C164112704 @default.
- W2909607453 hasConceptScore W2909607453C17744445 @default.
- W2909607453 hasConceptScore W2909607453C183115368 @default.
- W2909607453 hasConceptScore W2909607453C199539241 @default.
- W2909607453 hasConceptScore W2909607453C2779089604 @default.
- W2909607453 hasConceptScore W2909607453C38652104 @default.
- W2909607453 hasConceptScore W2909607453C41008148 @default.
- W2909607453 hasConceptScore W2909607453C41608201 @default.
- W2909607453 hasConceptScore W2909607453C50644808 @default.
- W2909607453 hasConceptScore W2909607453C65856478 @default.
- W2909607453 hasConceptScore W2909607453C71924100 @default.
- W2909607453 hasConceptScore W2909607453C95713431 @default.
- W2909607453 hasLocation W29096074531 @default.
- W2909607453 hasLocation W29096074532 @default.
- W2909607453 hasOpenAccess W2909607453 @default.
- W2909607453 hasPrimaryLocation W29096074531 @default.
- W2909607453 hasRelatedWork W1972662430 @default.
- W2909607453 hasRelatedWork W2150361697 @default.
- W2909607453 hasRelatedWork W2150608900 @default.
- W2909607453 hasRelatedWork W2553172991 @default.
- W2909607453 hasRelatedWork W4223943233 @default.
- W2909607453 hasRelatedWork W4312200629 @default.
- W2909607453 hasRelatedWork W4360585206 @default.
- W2909607453 hasRelatedWork W4364306694 @default.
- W2909607453 hasRelatedWork W4380075502 @default.
- W2909607453 hasRelatedWork W4380086463 @default.
- W2909607453 isParatext "false" @default.
- W2909607453 isRetracted "false" @default.
- W2909607453 magId "2909607453" @default.
- W2909607453 workType "article" @default.