Matches in SemOpenAlex for { <https://semopenalex.org/work/W2969908034> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W2969908034 abstract "Adversarial examples (AEs) are crafted by adding human-imperceptible perturbations to inputs such that a machine-learning based classifier incorrectly labels them. They have become a severe threat to the trustworthiness of machine learning. While AEs in the image domain have been well studied, audio AEs are less investigated. Recently, multiple techniques are proposed to generate audio AEs, which makes countermeasures against them urgent. Our experiments show that, given an audio AE, the transcription results by Automatic Speech Recognition (ASR) systems differ significantly (that is, poor transferability), as different ASR systems use different architectures, parameters, and training datasets. Based on this fact and inspired by Multiversion Programming, we propose a novel audio AE detection approach MVP-Ears, which utilizes the diverse off-the-shelf ASRs to determine whether an audio is an AE. We build the largest audio AE dataset to our knowledge, and the evaluation shows that the detection accuracy reaches 99.88%. While transferable audio AEs are difficult to generate at this moment, they may become a reality in future. We further adapt the idea above to proactively train the detection system for coping with transferable audio AEs. Thus, the proactive detection system is one giant step ahead of attackers working on transferable AEs." @default.
- W2969908034 created "2019-08-29" @default.
- W2969908034 creator A5018866702 @default.
- W2969908034 creator A5019793023 @default.
- W2969908034 creator A5024107992 @default.
- W2969908034 creator A5045089479 @default.
- W2969908034 creator A5060514022 @default.
- W2969908034 creator A5067262852 @default.
- W2969908034 creator A5083065490 @default.
- W2969908034 creator A5090491104 @default.
- W2969908034 date "2019-06-01" @default.
- W2969908034 modified "2023-09-23" @default.
- W2969908034 title "A Multiversion Programming Inspired Approach to Detecting Audio Adversarial Examples" @default.
- W2969908034 cites W2048060899 @default.
- W2969908034 cites W2064675550 @default.
- W2969908034 cites W2096330373 @default.
- W2969908034 cites W2127141656 @default.
- W2969908034 cites W2143612262 @default.
- W2969908034 cites W2180612164 @default.
- W2969908034 cites W2243397390 @default.
- W2969908034 cites W2397897814 @default.
- W2969908034 cites W2398932003 @default.
- W2969908034 cites W2735006420 @default.
- W2969908034 cites W2963408280 @default.
- W2969908034 cites W2963564844 @default.
- W2969908034 cites W2964301649 @default.
- W2969908034 doi "https://doi.org/10.1109/dsn.2019.00019" @default.
- W2969908034 hasPublicationYear "2019" @default.
- W2969908034 type Work @default.
- W2969908034 sameAs 2969908034 @default.
- W2969908034 citedByCount "27" @default.
- W2969908034 countsByYear W29699080342018 @default.
- W2969908034 countsByYear W29699080342019 @default.
- W2969908034 countsByYear W29699080342020 @default.
- W2969908034 countsByYear W29699080342021 @default.
- W2969908034 countsByYear W29699080342022 @default.
- W2969908034 countsByYear W29699080342023 @default.
- W2969908034 crossrefType "proceedings-article" @default.
- W2969908034 hasAuthorship W2969908034A5018866702 @default.
- W2969908034 hasAuthorship W2969908034A5019793023 @default.
- W2969908034 hasAuthorship W2969908034A5024107992 @default.
- W2969908034 hasAuthorship W2969908034A5045089479 @default.
- W2969908034 hasAuthorship W2969908034A5060514022 @default.
- W2969908034 hasAuthorship W2969908034A5067262852 @default.
- W2969908034 hasAuthorship W2969908034A5083065490 @default.
- W2969908034 hasAuthorship W2969908034A5090491104 @default.
- W2969908034 hasBestOaLocation W29699080342 @default.
- W2969908034 hasConcept C119857082 @default.
- W2969908034 hasConcept C127220857 @default.
- W2969908034 hasConcept C13895895 @default.
- W2969908034 hasConcept C140331021 @default.
- W2969908034 hasConcept C154945302 @default.
- W2969908034 hasConcept C28490314 @default.
- W2969908034 hasConcept C37736160 @default.
- W2969908034 hasConcept C41008148 @default.
- W2969908034 hasConcept C61272859 @default.
- W2969908034 hasConcept C64922751 @default.
- W2969908034 hasConcept C95623464 @default.
- W2969908034 hasConceptScore W2969908034C119857082 @default.
- W2969908034 hasConceptScore W2969908034C127220857 @default.
- W2969908034 hasConceptScore W2969908034C13895895 @default.
- W2969908034 hasConceptScore W2969908034C140331021 @default.
- W2969908034 hasConceptScore W2969908034C154945302 @default.
- W2969908034 hasConceptScore W2969908034C28490314 @default.
- W2969908034 hasConceptScore W2969908034C37736160 @default.
- W2969908034 hasConceptScore W2969908034C41008148 @default.
- W2969908034 hasConceptScore W2969908034C61272859 @default.
- W2969908034 hasConceptScore W2969908034C64922751 @default.
- W2969908034 hasConceptScore W2969908034C95623464 @default.
- W2969908034 hasLocation W29699080341 @default.
- W2969908034 hasLocation W29699080342 @default.
- W2969908034 hasOpenAccess W2969908034 @default.
- W2969908034 hasPrimaryLocation W29699080341 @default.
- W2969908034 hasRelatedWork W2357678230 @default.
- W2969908034 hasRelatedWork W2900076104 @default.
- W2969908034 hasRelatedWork W2961085424 @default.
- W2969908034 hasRelatedWork W2963030114 @default.
- W2969908034 hasRelatedWork W3081271317 @default.
- W2969908034 hasRelatedWork W3111953316 @default.
- W2969908034 hasRelatedWork W3128798311 @default.
- W2969908034 hasRelatedWork W4226402597 @default.
- W2969908034 hasRelatedWork W4287332144 @default.
- W2969908034 hasRelatedWork W4304891817 @default.
- W2969908034 isParatext "false" @default.
- W2969908034 isRetracted "false" @default.
- W2969908034 magId "2969908034" @default.
- W2969908034 workType "article" @default.