Matches in SemOpenAlex for { <https://semopenalex.org/work/W4311398608> ?p ?o ?g. }
- W4311398608 endingPage "906" @default.
- W4311398608 startingPage "882" @default.
- W4311398608 abstract "Recent efforts have shown that training data is not secured through the generalization and abstraction of algorithms. This vulnerability to the training data has been expressed through membership inference attacks that seek to discover the use of specific records within the training dataset of a model. Additionally, disparate membership inference attacks have been shown to achieve better accuracy compared with their macro attack counterparts. These disparate membership inference attacks use a pragmatic approach to attack individual, more vulnerable sub-sets of the data, such as underrepresented classes. While previous work in this field has explored model vulnerability to these attacks, this effort explores the vulnerability of datasets themselves to disparate membership inference attacks. This is accomplished through the development of a vulnerability-classification model that classifies datasets as vulnerable or secure to these attacks. To develop this model, a vulnerability-classification dataset is developed from over 100 datasets—including frequently cited datasets within the field. These datasets are described using a feature set of over 100 features and assigned labels developed from a combination of various modeling and attack strategies. By averaging the attack accuracy over 13 different modeling and attack strategies, the authors explore the vulnerabilities of the datasets themselves as opposed to a particular modeling or attack effort. The in-class observational distance, width ratio, and the proportion of discrete features are found to dominate the attributes defining dataset vulnerability to disparate membership inference attacks. These features are explored in deeper detail and used to develop exploratory methods for hardening these class-based sub-datasets against attacks showing preliminary mitigation success with combinations of feature reduction and class-balancing strategies." @default.
- W4311398608 created "2022-12-26" @default.
- W4311398608 creator A5021075936 @default.
- W4311398608 creator A5048930772 @default.
- W4311398608 creator A5087105219 @default.
- W4311398608 date "2022-12-14" @default.
- W4311398608 modified "2023-09-27" @default.
- W4311398608 title "An Understanding of the Vulnerability of Datasets to Disparate Membership Inference Attacks" @default.
- W4311398608 cites W1517567737 @default.
- W4311398608 cites W1604956938 @default.
- W4311398608 cites W1788633279 @default.
- W4311398608 cites W1969885422 @default.
- W4311398608 cites W1991856651 @default.
- W4311398608 cites W1995396954 @default.
- W4311398608 cites W2006251480 @default.
- W4311398608 cites W2010590614 @default.
- W4311398608 cites W2051267297 @default.
- W4311398608 cites W2070855716 @default.
- W4311398608 cites W2070896531 @default.
- W4311398608 cites W2072613728 @default.
- W4311398608 cites W2074112114 @default.
- W4311398608 cites W2085988980 @default.
- W4311398608 cites W2086685817 @default.
- W4311398608 cites W2088422930 @default.
- W4311398608 cites W2103459159 @default.
- W4311398608 cites W2123147099 @default.
- W4311398608 cites W2130486630 @default.
- W4311398608 cites W2135930857 @default.
- W4311398608 cites W2159024459 @default.
- W4311398608 cites W2165533158 @default.
- W4311398608 cites W2238226741 @default.
- W4311398608 cites W2319270064 @default.
- W4311398608 cites W2371937910 @default.
- W4311398608 cites W2535690855 @default.
- W4311398608 cites W2612932006 @default.
- W4311398608 cites W2751490707 @default.
- W4311398608 cites W2797800583 @default.
- W4311398608 cites W2801922753 @default.
- W4311398608 cites W2804411218 @default.
- W4311398608 cites W2811973125 @default.
- W4311398608 cites W2912023992 @default.
- W4311398608 cites W2918462293 @default.
- W4311398608 cites W2919491917 @default.
- W4311398608 cites W2962835266 @default.
- W4311398608 cites W2963378725 @default.
- W4311398608 cites W2964461202 @default.
- W4311398608 cites W2965527189 @default.
- W4311398608 cites W3004509394 @default.
- W4311398608 cites W3032338640 @default.
- W4311398608 cites W3096463949 @default.
- W4311398608 cites W3096692244 @default.
- W4311398608 cites W3159855141 @default.
- W4311398608 cites W3204683301 @default.
- W4311398608 cites W3211490561 @default.
- W4311398608 cites W4231202109 @default.
- W4311398608 doi "https://doi.org/10.3390/jcp2040045" @default.
- W4311398608 hasPublicationYear "2022" @default.
- W4311398608 type Work @default.
- W4311398608 citedByCount "0" @default.
- W4311398608 crossrefType "journal-article" @default.
- W4311398608 hasAuthorship W4311398608A5021075936 @default.
- W4311398608 hasAuthorship W4311398608A5048930772 @default.
- W4311398608 hasAuthorship W4311398608A5087105219 @default.
- W4311398608 hasBestOaLocation W43113986081 @default.
- W4311398608 hasConcept C119857082 @default.
- W4311398608 hasConcept C124101348 @default.
- W4311398608 hasConcept C134306372 @default.
- W4311398608 hasConcept C154945302 @default.
- W4311398608 hasConcept C177148314 @default.
- W4311398608 hasConcept C177264268 @default.
- W4311398608 hasConcept C199360897 @default.
- W4311398608 hasConcept C202444582 @default.
- W4311398608 hasConcept C2776214188 @default.
- W4311398608 hasConcept C2777212361 @default.
- W4311398608 hasConcept C33923547 @default.
- W4311398608 hasConcept C38652104 @default.
- W4311398608 hasConcept C41008148 @default.
- W4311398608 hasConcept C95713431 @default.
- W4311398608 hasConcept C9652623 @default.
- W4311398608 hasConceptScore W4311398608C119857082 @default.
- W4311398608 hasConceptScore W4311398608C124101348 @default.
- W4311398608 hasConceptScore W4311398608C134306372 @default.
- W4311398608 hasConceptScore W4311398608C154945302 @default.
- W4311398608 hasConceptScore W4311398608C177148314 @default.
- W4311398608 hasConceptScore W4311398608C177264268 @default.
- W4311398608 hasConceptScore W4311398608C199360897 @default.
- W4311398608 hasConceptScore W4311398608C202444582 @default.
- W4311398608 hasConceptScore W4311398608C2776214188 @default.
- W4311398608 hasConceptScore W4311398608C2777212361 @default.
- W4311398608 hasConceptScore W4311398608C33923547 @default.
- W4311398608 hasConceptScore W4311398608C38652104 @default.
- W4311398608 hasConceptScore W4311398608C41008148 @default.
- W4311398608 hasConceptScore W4311398608C95713431 @default.
- W4311398608 hasConceptScore W4311398608C9652623 @default.
- W4311398608 hasIssue "4" @default.
- W4311398608 hasLocation W43113986081 @default.
- W4311398608 hasLocation W43113986082 @default.
- W4311398608 hasOpenAccess W4311398608 @default.