Matches in SemOpenAlex for { <https://semopenalex.org/work/W2783155279> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W2783155279 endingPage "1177" @default.
- W2783155279 startingPage "1164" @default.
- W2783155279 abstract "Adversarial learning is the study of machine learning techniques deployed in non-benign environments. Example applications include classification for detecting spam, network intrusion detection, and credit card scoring. In fact, as the use of machine learning grows in diverse application domains, the possibility for adversarial behavior is likely to increase. When adversarial learning is modelled in a game-theoretic setup, the standard assumption about the adversary (player) behavior is the ability to change all features of the classifiers (the opponent player) at will. The adversary pays a cost proportional to the size of the “attack”. We refer to this form of adversarial behavior as a dense feature attack. However, the aim of an adversary is not just to subvert a classifier but carry out data transformation in a way such that spam continues to remain effective. We demonstrate that an adversary could potentially achieve this objective by carrying out a sparse feature attack. We design an algorithm to show how a classifier should be designed to be robust against sparse adversarial attacks. Our main insight is that sparse feature attacks are best defended by designing classifiers which use ℓ <sub xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>1</sub> regularizers." @default.
- W2783155279 created "2018-01-26" @default.
- W2783155279 creator A5000050561 @default.
- W2783155279 creator A5037947876 @default.
- W2783155279 creator A5071037763 @default.
- W2783155279 creator A5086529957 @default.
- W2783155279 date "2018-06-01" @default.
- W2783155279 modified "2023-10-17" @default.
- W2783155279 title "Sparse Feature Attacks in Adversarial Learning" @default.
- W2783155279 cites W1965052658 @default.
- W2783155279 cites W1966912382 @default.
- W2783155279 cites W1990676004 @default.
- W2783155279 cites W1994520254 @default.
- W2783155279 cites W1996234057 @default.
- W2783155279 cites W2030611346 @default.
- W2783155279 cites W2094100186 @default.
- W2783155279 cites W2109300365 @default.
- W2783155279 cites W2114296159 @default.
- W2783155279 cites W2117173631 @default.
- W2783155279 cites W2145901875 @default.
- W2783155279 cites W2146211964 @default.
- W2783155279 cites W2293768274 @default.
- W2783155279 cites W2296452361 @default.
- W2783155279 cites W2401760721 @default.
- W2783155279 cites W2494395359 @default.
- W2783155279 cites W2787894218 @default.
- W2783155279 cites W4302571648 @default.
- W2783155279 doi "https://doi.org/10.1109/tkde.2018.2790928" @default.
- W2783155279 hasPublicationYear "2018" @default.
- W2783155279 type Work @default.
- W2783155279 sameAs 2783155279 @default.
- W2783155279 citedByCount "18" @default.
- W2783155279 countsByYear W27831552792019 @default.
- W2783155279 countsByYear W27831552792020 @default.
- W2783155279 countsByYear W27831552792021 @default.
- W2783155279 countsByYear W27831552792022 @default.
- W2783155279 countsByYear W27831552792023 @default.
- W2783155279 crossrefType "journal-article" @default.
- W2783155279 hasAuthorship W2783155279A5000050561 @default.
- W2783155279 hasAuthorship W2783155279A5037947876 @default.
- W2783155279 hasAuthorship W2783155279A5071037763 @default.
- W2783155279 hasAuthorship W2783155279A5086529957 @default.
- W2783155279 hasConcept C119857082 @default.
- W2783155279 hasConcept C138885662 @default.
- W2783155279 hasConcept C154945302 @default.
- W2783155279 hasConcept C2776401178 @default.
- W2783155279 hasConcept C2778403875 @default.
- W2783155279 hasConcept C35525427 @default.
- W2783155279 hasConcept C37736160 @default.
- W2783155279 hasConcept C38652104 @default.
- W2783155279 hasConcept C41008148 @default.
- W2783155279 hasConcept C41065033 @default.
- W2783155279 hasConcept C41895202 @default.
- W2783155279 hasConcept C541664917 @default.
- W2783155279 hasConcept C95623464 @default.
- W2783155279 hasConceptScore W2783155279C119857082 @default.
- W2783155279 hasConceptScore W2783155279C138885662 @default.
- W2783155279 hasConceptScore W2783155279C154945302 @default.
- W2783155279 hasConceptScore W2783155279C2776401178 @default.
- W2783155279 hasConceptScore W2783155279C2778403875 @default.
- W2783155279 hasConceptScore W2783155279C35525427 @default.
- W2783155279 hasConceptScore W2783155279C37736160 @default.
- W2783155279 hasConceptScore W2783155279C38652104 @default.
- W2783155279 hasConceptScore W2783155279C41008148 @default.
- W2783155279 hasConceptScore W2783155279C41065033 @default.
- W2783155279 hasConceptScore W2783155279C41895202 @default.
- W2783155279 hasConceptScore W2783155279C541664917 @default.
- W2783155279 hasConceptScore W2783155279C95623464 @default.
- W2783155279 hasIssue "6" @default.
- W2783155279 hasLocation W27831552791 @default.
- W2783155279 hasOpenAccess W2783155279 @default.
- W2783155279 hasPrimaryLocation W27831552791 @default.
- W2783155279 hasRelatedWork W2891797126 @default.
- W2783155279 hasRelatedWork W2921786109 @default.
- W2783155279 hasRelatedWork W3013617128 @default.
- W2783155279 hasRelatedWork W3046843850 @default.
- W2783155279 hasRelatedWork W3124408655 @default.
- W2783155279 hasRelatedWork W3216063557 @default.
- W2783155279 hasRelatedWork W4211047985 @default.
- W2783155279 hasRelatedWork W4233678362 @default.
- W2783155279 hasRelatedWork W4286899967 @default.
- W2783155279 hasRelatedWork W4386716251 @default.
- W2783155279 hasVolume "30" @default.
- W2783155279 isParatext "false" @default.
- W2783155279 isRetracted "false" @default.
- W2783155279 magId "2783155279" @default.
- W2783155279 workType "article" @default.