Matches in SemOpenAlex for { <https://semopenalex.org/work/W2952541330> ?p ?o ?g. }
- W2952541330 abstract "Machine learning algorithms are vulnerable to poisoning attacks: An adversary can inject malicious points in the training dataset to influence the learning process and degrade the algorithm's performance. Optimal poisoning attacks have already been proposed to evaluate worst-case scenarios, modelling attacks as a bi-level optimization problem. Solving these problems is computationally demanding and has limited applicability for some models such as deep networks. In this paper we introduce a novel generative model to craft systematic poisoning attacks against machine learning classifiers generating adversarial training examples, i.e. samples that look like genuine data points but that degrade the classifier's accuracy when used for training. We propose a Generative Adversarial Net with three components: generator, discriminator, and the target classifier. This approach allows us to model naturally the detectability constrains that can be expected in realistic attacks and to identify the regions of the underlying data distribution that can be more vulnerable to data poisoning. Our experimental evaluation shows the effectiveness of our attack to compromise machine learning classifiers, including deep networks." @default.
- W2952541330 created "2019-06-27" @default.
- W2952541330 creator A5001385652 @default.
- W2952541330 creator A5026216647 @default.
- W2952541330 creator A5032718837 @default.
- W2952541330 creator A5034656746 @default.
- W2952541330 creator A5065619732 @default.
- W2952541330 date "2019-09-25" @default.
- W2952541330 modified "2023-09-23" @default.
- W2952541330 title "Poisoning Attacks with Generative Adversarial Nets." @default.
- W2952541330 cites W1492810482 @default.
- W2952541330 cites W2095577883 @default.
- W2952541330 cites W2099471712 @default.
- W2952541330 cites W2112507308 @default.
- W2952541330 cites W2112796928 @default.
- W2952541330 cites W2125389028 @default.
- W2952541330 cites W2125908420 @default.
- W2952541330 cites W2134063365 @default.
- W2952541330 cites W2151298633 @default.
- W2952541330 cites W2162552722 @default.
- W2952541330 cites W2167421362 @default.
- W2952541330 cites W2170651405 @default.
- W2952541330 cites W2293844262 @default.
- W2952541330 cites W2541159141 @default.
- W2952541330 cites W2591602089 @default.
- W2952541330 cites W2597603852 @default.
- W2952541330 cites W2748789698 @default.
- W2952541330 cites W2750384547 @default.
- W2952541330 cites W2783555701 @default.
- W2952541330 cites W2785509559 @default.
- W2952541330 cites W2898998737 @default.
- W2952541330 cites W2962763344 @default.
- W2952541330 cites W2963207607 @default.
- W2952541330 cites W2963343288 @default.
- W2952541330 cites W2963373786 @default.
- W2952541330 cites W2963618920 @default.
- W2952541330 cites W2963888996 @default.
- W2952541330 cites W3118608800 @default.
- W2952541330 hasPublicationYear "2019" @default.
- W2952541330 type Work @default.
- W2952541330 sameAs 2952541330 @default.
- W2952541330 citedByCount "13" @default.
- W2952541330 countsByYear W29525413302018 @default.
- W2952541330 countsByYear W29525413302019 @default.
- W2952541330 countsByYear W29525413302020 @default.
- W2952541330 countsByYear W29525413302021 @default.
- W2952541330 crossrefType "posted-content" @default.
- W2952541330 hasAuthorship W2952541330A5001385652 @default.
- W2952541330 hasAuthorship W2952541330A5026216647 @default.
- W2952541330 hasAuthorship W2952541330A5032718837 @default.
- W2952541330 hasAuthorship W2952541330A5034656746 @default.
- W2952541330 hasAuthorship W2952541330A5065619732 @default.
- W2952541330 hasConcept C108583219 @default.
- W2952541330 hasConcept C119857082 @default.
- W2952541330 hasConcept C154945302 @default.
- W2952541330 hasConcept C2778403875 @default.
- W2952541330 hasConcept C2779803651 @default.
- W2952541330 hasConcept C2988773926 @default.
- W2952541330 hasConcept C37736160 @default.
- W2952541330 hasConcept C38652104 @default.
- W2952541330 hasConcept C39890363 @default.
- W2952541330 hasConcept C41008148 @default.
- W2952541330 hasConcept C41065033 @default.
- W2952541330 hasConcept C76155785 @default.
- W2952541330 hasConcept C94915269 @default.
- W2952541330 hasConcept C95623464 @default.
- W2952541330 hasConceptScore W2952541330C108583219 @default.
- W2952541330 hasConceptScore W2952541330C119857082 @default.
- W2952541330 hasConceptScore W2952541330C154945302 @default.
- W2952541330 hasConceptScore W2952541330C2778403875 @default.
- W2952541330 hasConceptScore W2952541330C2779803651 @default.
- W2952541330 hasConceptScore W2952541330C2988773926 @default.
- W2952541330 hasConceptScore W2952541330C37736160 @default.
- W2952541330 hasConceptScore W2952541330C38652104 @default.
- W2952541330 hasConceptScore W2952541330C39890363 @default.
- W2952541330 hasConceptScore W2952541330C41008148 @default.
- W2952541330 hasConceptScore W2952541330C41065033 @default.
- W2952541330 hasConceptScore W2952541330C76155785 @default.
- W2952541330 hasConceptScore W2952541330C94915269 @default.
- W2952541330 hasConceptScore W2952541330C95623464 @default.
- W2952541330 hasLocation W29525413301 @default.
- W2952541330 hasOpenAccess W2952541330 @default.
- W2952541330 hasPrimaryLocation W29525413301 @default.
- W2952541330 hasRelatedWork W2117539524 @default.
- W2952541330 hasRelatedWork W2167421362 @default.
- W2952541330 hasRelatedWork W2293844262 @default.
- W2952541330 hasRelatedWork W2591602089 @default.
- W2952541330 hasRelatedWork W2597603852 @default.
- W2952541330 hasRelatedWork W2748789698 @default.
- W2952541330 hasRelatedWork W2774423163 @default.
- W2952541330 hasRelatedWork W2785509559 @default.
- W2952541330 hasRelatedWork W2898998737 @default.
- W2952541330 hasRelatedWork W2946227741 @default.
- W2952541330 hasRelatedWork W2949506549 @default.
- W2952541330 hasRelatedWork W2963343288 @default.
- W2952541330 hasRelatedWork W2963618920 @default.
- W2952541330 hasRelatedWork W2963888996 @default.
- W2952541330 hasRelatedWork W2964043980 @default.
- W2952541330 hasRelatedWork W2964153729 @default.
- W2952541330 hasRelatedWork W2972906846 @default.