Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197041852> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W3197041852 endingPage "48" @default.
- W3197041852 startingPage "35" @default.
- W3197041852 abstract "Neural Networks are used today in numerous security- and safety-relevant domains and are, as such, a popular target of attacks that subvert their classification capabilities, by manipulating the network parameters. Prior work has introduced sensitive samples -- inputs highly sensitive to parameter changes -- to detect such manipulations, and proposed a gradient ascent-based approach to compute them. In this paper we offer an alternative, using symbolic constraint solvers. We model the network and a formal specification of a sensitive sample in the language of the solver and ask for a solution. This approach supports a rich class of queries, corresponding, for instance, to the presence of certain types of attacks. Unlike earlier techniques, our approach does not depend on convex search domains, or on the suitability of a starting point for the search. We address the performance limitations of constraint solvers by partitioning the search space for the solver, and exploring the partitions according to a balanced schedule that still retains completeness of the search. We demonstrate the impact of the use of solvers in terms of functionality and search efficiency, using a case study for the detection of Trojan attacks on Neural Networks." @default.
- W3197041852 created "2021-09-13" @default.
- W3197041852 creator A5023675966 @default.
- W3197041852 creator A5025192161 @default.
- W3197041852 creator A5064729877 @default.
- W3197041852 creator A5083131515 @default.
- W3197041852 date "2021-09-07" @default.
- W3197041852 modified "2023-09-26" @default.
- W3197041852 title "Sensitive Samples Revisited: Detecting Neural Network Attacks Using Constraint Solvers" @default.
- W3197041852 cites W1486132939 @default.
- W3197041852 cites W1977610018 @default.
- W3197041852 cites W2258415152 @default.
- W3197041852 cites W2594877703 @default.
- W3197041852 cites W2753783305 @default.
- W3197041852 cites W2934843808 @default.
- W3197041852 cites W2946149952 @default.
- W3197041852 cites W2948833786 @default.
- W3197041852 cites W2966689772 @default.
- W3197041852 cites W2990270730 @default.
- W3197041852 cites W3041840141 @default.
- W3197041852 doi "https://doi.org/10.4204/eptcs.342.4" @default.
- W3197041852 hasPublicationYear "2021" @default.
- W3197041852 type Work @default.
- W3197041852 sameAs 3197041852 @default.
- W3197041852 citedByCount "0" @default.
- W3197041852 crossrefType "journal-article" @default.
- W3197041852 hasAuthorship W3197041852A5023675966 @default.
- W3197041852 hasAuthorship W3197041852A5025192161 @default.
- W3197041852 hasAuthorship W3197041852A5064729877 @default.
- W3197041852 hasAuthorship W3197041852A5083131515 @default.
- W3197041852 hasBestOaLocation W31970418521 @default.
- W3197041852 hasConcept C119857082 @default.
- W3197041852 hasConcept C134306372 @default.
- W3197041852 hasConcept C154945302 @default.
- W3197041852 hasConcept C17231256 @default.
- W3197041852 hasConcept C174333608 @default.
- W3197041852 hasConcept C199360897 @default.
- W3197041852 hasConcept C199622910 @default.
- W3197041852 hasConcept C2524010 @default.
- W3197041852 hasConcept C2776036281 @default.
- W3197041852 hasConcept C2777212361 @default.
- W3197041852 hasConcept C2778770139 @default.
- W3197041852 hasConcept C28719098 @default.
- W3197041852 hasConcept C33923547 @default.
- W3197041852 hasConcept C38652104 @default.
- W3197041852 hasConcept C41008148 @default.
- W3197041852 hasConcept C49937458 @default.
- W3197041852 hasConcept C50644808 @default.
- W3197041852 hasConcept C80444323 @default.
- W3197041852 hasConceptScore W3197041852C119857082 @default.
- W3197041852 hasConceptScore W3197041852C134306372 @default.
- W3197041852 hasConceptScore W3197041852C154945302 @default.
- W3197041852 hasConceptScore W3197041852C17231256 @default.
- W3197041852 hasConceptScore W3197041852C174333608 @default.
- W3197041852 hasConceptScore W3197041852C199360897 @default.
- W3197041852 hasConceptScore W3197041852C199622910 @default.
- W3197041852 hasConceptScore W3197041852C2524010 @default.
- W3197041852 hasConceptScore W3197041852C2776036281 @default.
- W3197041852 hasConceptScore W3197041852C2777212361 @default.
- W3197041852 hasConceptScore W3197041852C2778770139 @default.
- W3197041852 hasConceptScore W3197041852C28719098 @default.
- W3197041852 hasConceptScore W3197041852C33923547 @default.
- W3197041852 hasConceptScore W3197041852C38652104 @default.
- W3197041852 hasConceptScore W3197041852C41008148 @default.
- W3197041852 hasConceptScore W3197041852C49937458 @default.
- W3197041852 hasConceptScore W3197041852C50644808 @default.
- W3197041852 hasConceptScore W3197041852C80444323 @default.
- W3197041852 hasLocation W31970418521 @default.
- W3197041852 hasLocation W31970418522 @default.
- W3197041852 hasOpenAccess W3197041852 @default.
- W3197041852 hasPrimaryLocation W31970418521 @default.
- W3197041852 hasRelatedWork W1499132241 @default.
- W3197041852 hasRelatedWork W1581054293 @default.
- W3197041852 hasRelatedWork W1610655459 @default.
- W3197041852 hasRelatedWork W1848884953 @default.
- W3197041852 hasRelatedWork W1975963464 @default.
- W3197041852 hasRelatedWork W2046194288 @default.
- W3197041852 hasRelatedWork W2953287580 @default.
- W3197041852 hasRelatedWork W4300784158 @default.
- W3197041852 hasRelatedWork W4301765873 @default.
- W3197041852 hasRelatedWork W1629725936 @default.
- W3197041852 hasVolume "342" @default.
- W3197041852 isParatext "false" @default.
- W3197041852 isRetracted "false" @default.
- W3197041852 magId "3197041852" @default.
- W3197041852 workType "article" @default.