Matches in SemOpenAlex for { <https://semopenalex.org/work/W3038965404> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W3038965404 abstract "As the machine learning technology is being widely used in next-generation networks aided by edge computing, its security is of great concern for network maintenance and management. On the other hand, intelligent data processing techniques for enhancing the performance of next-generation networks are mainly used to reinforce resource management, control and optimization. Therefore, compared to learning a classifier for other application scenarios (e.g., pattern recognition, network intrusion detection, image identification, etc.), learning a good regressor via machine learning is more valuable for edge computing applications in next-generation networks. Hence, we revisit the security of polynomial regression learning in this paper and propose a scalable poisoning method against regression-type edge computing applications via an approximate optimization strategy. Specifically, we first analyze the existing gradientbased optimization poisoning method (called OptP) against linear regression model. Then, we extend the method and give a formal description of poisoning against polynomial regression with order greater than one. After that, we propose a greedy optimization strategy to obtain good adversarial examples for compromising polynomial regression-type applications. Finally, we demonstrate the superior performance of the proposed poisoning method by applying it to attack against two synthetic data sets and three public data sets from three different application scenarios." @default.
- W3038965404 created "2020-07-10" @default.
- W3038965404 creator A5037159442 @default.
- W3038965404 creator A5070386973 @default.
- W3038965404 creator A5076409244 @default.
- W3038965404 creator A5077798968 @default.
- W3038965404 creator A5088213590 @default.
- W3038965404 date "2019-04-29" @default.
- W3038965404 modified "2023-09-23" @default.
- W3038965404 title "Scalable Poisoning Against Regression-Type Edge Computing Applications via An Approximate Optimization Strategy" @default.
- W3038965404 cites W1977367431 @default.
- W3038965404 cites W2004255221 @default.
- W3038965404 cites W2040340473 @default.
- W3038965404 cites W2094547360 @default.
- W3038965404 cites W2293387016 @default.
- W3038965404 cites W2293844262 @default.
- W3038965404 cites W2319556062 @default.
- W3038965404 cites W2730315335 @default.
- W3038965404 cites W2791319131 @default.
- W3038965404 cites W2897798676 @default.
- W3038965404 cites W2962763344 @default.
- W3038965404 cites W3120740533 @default.
- W3038965404 hasPublicationYear "2019" @default.
- W3038965404 type Work @default.
- W3038965404 sameAs 3038965404 @default.
- W3038965404 citedByCount "0" @default.
- W3038965404 crossrefType "proceedings-article" @default.
- W3038965404 hasAuthorship W3038965404A5037159442 @default.
- W3038965404 hasAuthorship W3038965404A5070386973 @default.
- W3038965404 hasAuthorship W3038965404A5076409244 @default.
- W3038965404 hasAuthorship W3038965404A5077798968 @default.
- W3038965404 hasAuthorship W3038965404A5088213590 @default.
- W3038965404 hasConcept C11413529 @default.
- W3038965404 hasConcept C119857082 @default.
- W3038965404 hasConcept C120068334 @default.
- W3038965404 hasConcept C124101348 @default.
- W3038965404 hasConcept C137836250 @default.
- W3038965404 hasConcept C152877465 @default.
- W3038965404 hasConcept C154945302 @default.
- W3038965404 hasConcept C41008148 @default.
- W3038965404 hasConcept C48044578 @default.
- W3038965404 hasConcept C77088390 @default.
- W3038965404 hasConcept C95623464 @default.
- W3038965404 hasConceptScore W3038965404C11413529 @default.
- W3038965404 hasConceptScore W3038965404C119857082 @default.
- W3038965404 hasConceptScore W3038965404C120068334 @default.
- W3038965404 hasConceptScore W3038965404C124101348 @default.
- W3038965404 hasConceptScore W3038965404C137836250 @default.
- W3038965404 hasConceptScore W3038965404C152877465 @default.
- W3038965404 hasConceptScore W3038965404C154945302 @default.
- W3038965404 hasConceptScore W3038965404C41008148 @default.
- W3038965404 hasConceptScore W3038965404C48044578 @default.
- W3038965404 hasConceptScore W3038965404C77088390 @default.
- W3038965404 hasConceptScore W3038965404C95623464 @default.
- W3038965404 hasLocation W30389654041 @default.
- W3038965404 hasOpenAccess W3038965404 @default.
- W3038965404 hasPrimaryLocation W30389654041 @default.
- W3038965404 hasRelatedWork W1544234128 @default.
- W3038965404 hasRelatedWork W1565090700 @default.
- W3038965404 hasRelatedWork W1596802713 @default.
- W3038965404 hasRelatedWork W2038361169 @default.
- W3038965404 hasRelatedWork W2043524669 @default.
- W3038965404 hasRelatedWork W2282094334 @default.
- W3038965404 hasRelatedWork W2293865981 @default.
- W3038965404 hasRelatedWork W2734573782 @default.
- W3038965404 hasRelatedWork W2786726542 @default.
- W3038965404 hasRelatedWork W2809263317 @default.
- W3038965404 hasRelatedWork W2888764088 @default.
- W3038965404 hasRelatedWork W2913324419 @default.
- W3038965404 hasRelatedWork W3040845823 @default.
- W3038965404 hasRelatedWork W3100126642 @default.
- W3038965404 hasRelatedWork W3100520057 @default.
- W3038965404 hasRelatedWork W3100677605 @default.
- W3038965404 hasRelatedWork W3112837374 @default.
- W3038965404 hasRelatedWork W3124270557 @default.
- W3038965404 hasRelatedWork W3174903366 @default.
- W3038965404 hasRelatedWork W3198908542 @default.
- W3038965404 isParatext "false" @default.
- W3038965404 isRetracted "false" @default.
- W3038965404 magId "3038965404" @default.
- W3038965404 workType "article" @default.