FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200942016> ?p ?o ?g. }

• W3200942016 abstract "Machine learning researchers have long noticed the phenomenon that the model training process will be more effective and efficient when the training samples are densely sampled around the underlying decision boundary. While this observation has already been widely applied in a range of machine learning security techniques, it lacks theoretical analyses of the correctness of the observation. To address this challenge, we first add particular perturbation to original training examples using adversarial attack methods so that the generated examples could lie approximately on the decision boundary of the ML classifiers. We then investigate the connections between active learning and these particular training examples. Through analyzing various representative classifiers such as k-NN classifiers, kernel methods as well as deep neural networks, we establish a theoretical foundation for the observation. As a result, our theoretical proofs provide support to more efficient active learning methods with the help of adversarial examples, contrary to previous works where adversarial examples are often used as destructive solutions. Experimental results show that the established theoretical foundation will guide better active learning strategies based on adversarial examples." @default.
• W3200942016 created "2021-09-27" @default.
• W3200942016 creator A5009635380 @default.
• W3200942016 creator A5017464942 @default.
• W3200942016 creator A5033785543 @default.
• W3200942016 creator A5068080901 @default.
• W3200942016 creator A5090083857 @default.
• W3200942016 date "2021-09-22" @default.
• W3200942016 modified "2023-09-23" @default.
• W3200942016 title "Exploring Adversarial Examples for Efficient Active Learning in Machine Learning Classifiers" @default.
• W3200942016 cites W1673923490 @default.
• W3200942016 cites W1885172175 @default.
• W3200942016 cites W1994877143 @default.
• W3200942016 cites W2012501405 @default.
• W3200942016 cites W2027057364 @default.
• W3200942016 cites W2073898607 @default.
• W3200942016 cites W2101234009 @default.
• W3200942016 cites W2108882154 @default.
• W3200942016 cites W2113037782 @default.
• W3200942016 cites W2128678390 @default.
• W3200942016 cites W2171671120 @default.
• W3200942016 cites W2180612164 @default.
• W3200942016 cites W2243397390 @default.
• W3200942016 cites W2303810353 @default.
• W3200942016 cites W2408141691 @default.
• W3200942016 cites W2413904250 @default.
• W3200942016 cites W2426031434 @default.
• W3200942016 cites W2535873859 @default.
• W3200942016 cites W2570764145 @default.
• W3200942016 cites W2603766943 @default.
• W3200942016 cites W2603775267 @default.
• W3200942016 cites W2760947028 @default.
• W3200942016 cites W2773446523 @default.
• W3200942016 cites W2788686132 @default.
• W3200942016 cites W2805918584 @default.
• W3200942016 cites W2874797877 @default.
• W3200942016 cites W2895160027 @default.
• W3200942016 cites W2897865027 @default.
• W3200942016 cites W2948510860 @default.
• W3200942016 cites W2953941229 @default.
• W3200942016 cites W2963207607 @default.
• W3200942016 cites W2963539275 @default.
• W3200942016 cites W2963857521 @default.
• W3200942016 cites W2964082701 @default.
• W3200942016 cites W2964149421 @default.
• W3200942016 cites W2964220529 @default.
• W3200942016 cites W2964253222 @default.
• W3200942016 cites W3007318395 @default.
• W3200942016 cites W3018424040 @default.
• W3200942016 cites W3033355066 @default.
• W3200942016 cites W3037012516 @default.
• W3200942016 cites W3049515540 @default.
• W3200942016 cites W3103790796 @default.
• W3200942016 cites W3118608800 @default.
• W3200942016 cites W3120740533 @default.
• W3200942016 doi "https://doi.org/10.48550/arxiv.2109.10770" @default.
• W3200942016 hasPublicationYear "2021" @default.
• W3200942016 type Work @default.
• W3200942016 sameAs 3200942016 @default.
• W3200942016 citedByCount "0" @default.
• W3200942016 crossrefType "posted-content" @default.
• W3200942016 hasAuthorship W3200942016A5009635380 @default.
• W3200942016 hasAuthorship W3200942016A5017464942 @default.
• W3200942016 hasAuthorship W3200942016A5033785543 @default.
• W3200942016 hasAuthorship W3200942016A5068080901 @default.
• W3200942016 hasAuthorship W3200942016A5090083857 @default.
• W3200942016 hasBestOaLocation W32009420161 @default.
• W3200942016 hasConcept C108583219 @default.
• W3200942016 hasConcept C11413529 @default.
• W3200942016 hasConcept C119857082 @default.
• W3200942016 hasConcept C154945302 @default.
• W3200942016 hasConcept C24138899 @default.
• W3200942016 hasConcept C2778403875 @default.
• W3200942016 hasConcept C37736160 @default.
• W3200942016 hasConcept C41008148 @default.
• W3200942016 hasConcept C42023084 @default.
• W3200942016 hasConcept C50644808 @default.
• W3200942016 hasConcept C55439883 @default.
• W3200942016 hasConcept C77967617 @default.
• W3200942016 hasConcept C95623464 @default.
• W3200942016 hasConceptScore W3200942016C108583219 @default.
• W3200942016 hasConceptScore W3200942016C11413529 @default.
• W3200942016 hasConceptScore W3200942016C119857082 @default.
• W3200942016 hasConceptScore W3200942016C154945302 @default.
• W3200942016 hasConceptScore W3200942016C24138899 @default.
• W3200942016 hasConceptScore W3200942016C2778403875 @default.
• W3200942016 hasConceptScore W3200942016C37736160 @default.
• W3200942016 hasConceptScore W3200942016C41008148 @default.
• W3200942016 hasConceptScore W3200942016C42023084 @default.
• W3200942016 hasConceptScore W3200942016C50644808 @default.
• W3200942016 hasConceptScore W3200942016C55439883 @default.
• W3200942016 hasConceptScore W3200942016C77967617 @default.
• W3200942016 hasConceptScore W3200942016C95623464 @default.
• W3200942016 hasLocation W32009420161 @default.
• W3200942016 hasOpenAccess W3200942016 @default.
• W3200942016 hasPrimaryLocation W32009420161 @default.
• W3200942016 hasRelatedWork W2597787948 @default.
• W3200942016 hasRelatedWork W2941584439 @default.
• W3200942016 hasRelatedWork W2951786554 @default.
• W3200942016 hasRelatedWork W3040460291 @default.

• next