FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3083544013> ?p ?o ?g. }

• W3083544013 abstract "Post-hoc explanation techniques refer to a posteriori methods that can be used to explain how black-box machine learning models produce their outcomes. Among post-hoc explanation techniques, counterfactual explanations are becoming one of the most popular methods to achieve this objective. In particular, in addition to highlighting the most important features used by the black-box model, they provide users with actionable explanations in the form of data instances that would have received a different outcome. Nonetheless, by doing so, they also leak non-trivial information about the model itself, which raises privacy issues. In this work, we demonstrate how an adversary can leverage the information provided by counterfactual explanations to build high-fidelity and high-accuracy model extraction attacks. More precisely, our attack enables the adversary to build a faithful copy of a target model by accessing its counterfactual explanations. The empirical evaluation of the proposed attack on black-box models trained on real-world datasets demonstrates that they can achieve high-fidelity and high-accuracy extraction even under low query budgets." @default.
• W3083544013 created "2020-09-11" @default.
• W3083544013 creator A5000861121 @default.
• W3083544013 creator A5022365448 @default.
• W3083544013 creator A5049604013 @default.
• W3083544013 date "2020-09-03" @default.
• W3083544013 modified "2023-09-27" @default.
• W3083544013 title "Model extraction from counterfactual explanations." @default.
• W3083544013 cites W1473189865 @default.
• W3083544013 cites W1503398984 @default.
• W3083544013 cites W1873763122 @default.
• W3083544013 cites W2002303652 @default.
• W3083544013 cites W2048679005 @default.
• W3083544013 cites W2051267297 @default.
• W3083544013 cites W2109426455 @default.
• W3083544013 cites W2113882472 @default.
• W3083544013 cites W2139709458 @default.
• W3083544013 cites W2150165932 @default.
• W3083544013 cites W2154565402 @default.
• W3083544013 cites W2158585626 @default.
• W3083544013 cites W2282821441 @default.
• W3083544013 cites W2286670277 @default.
• W3083544013 cites W2296452361 @default.
• W3083544013 cites W2394669110 @default.
• W3083544013 cites W2428981601 @default.
• W3083544013 cites W2493343568 @default.
• W3083544013 cites W2532781556 @default.
• W3083544013 cites W2551317447 @default.
• W3083544013 cites W2551974706 @default.
• W3083544013 cites W2603766943 @default.
• W3083544013 cites W2657631929 @default.
• W3083544013 cites W2744365997 @default.
• W3083544013 cites W2750144484 @default.
• W3083544013 cites W2777473055 @default.
• W3083544013 cites W2804657206 @default.
• W3083544013 cites W2807992309 @default.
• W3083544013 cites W2811276992 @default.
• W3083544013 cites W2891003389 @default.
• W3083544013 cites W2891612330 @default.
• W3083544013 cites W2901277930 @default.
• W3083544013 cites W2909392392 @default.
• W3083544013 cites W2911964244 @default.
• W3083544013 cites W2914635529 @default.
• W3083544013 cites W2944977718 @default.
• W3083544013 cites W2945976633 @default.
• W3083544013 cites W2947527202 @default.
• W3083544013 cites W2951306478 @default.
• W3083544013 cites W2953522645 @default.
• W3083544013 cites W2954172636 @default.
• W3083544013 cites W2954266614 @default.
• W3083544013 cites W2962772482 @default.
• W3083544013 cites W2962790223 @default.
• W3083544013 cites W2962807381 @default.
• W3083544013 cites W2962843949 @default.
• W3083544013 cites W2962851944 @default.
• W3083544013 cites W2962862931 @default.
• W3083544013 cites W2962966435 @default.
• W3083544013 cites W2963207607 @default.
• W3083544013 cites W2963303354 @default.
• W3083544013 cites W2963465081 @default.
• W3083544013 cites W2963560987 @default.
• W3083544013 cites W2963844355 @default.
• W3083544013 cites W2964112969 @default.
• W3083544013 cites W2964134873 @default.
• W3083544013 cites W2964153729 @default.
• W3083544013 cites W2964212578 @default.
• W3083544013 cites W2964318098 @default.
• W3083544013 cites W2964449086 @default.
• W3083544013 cites W2969695741 @default.
• W3083544013 cites W2970658946 @default.
• W3083544013 cites W2972216789 @default.
• W3083544013 cites W2978126126 @default.
• W3083544013 cites W2979320098 @default.
• W3083544013 cites W2984851138 @default.
• W3083544013 cites W2994056986 @default.
• W3083544013 cites W2997428643 @default.
• W3083544013 cites W3004315562 @default.
• W3083544013 cites W3005073185 @default.
• W3083544013 cites W3007665694 @default.
• W3083544013 cites W3018424040 @default.
• W3083544013 cites W3022179901 @default.
• W3083544013 cites W3034710707 @default.
• W3083544013 cites W3037724138 @default.
• W3083544013 cites W3042265470 @default.
• W3083544013 cites W3049515540 @default.
• W3083544013 cites W3098155791 @default.
• W3083544013 cites W3102161834 @default.
• W3083544013 cites W3102834905 @default.
• W3083544013 cites W3120740533 @default.
• W3083544013 cites W3122175177 @default.
• W3083544013 cites W359283280 @default.
• W3083544013 cites W51263609 @default.
• W3083544013 hasPublicationYear "2020" @default.
• W3083544013 type Work @default.
• W3083544013 sameAs 3083544013 @default.
• W3083544013 citedByCount "6" @default.
• W3083544013 countsByYear W30835440132019 @default.
• W3083544013 countsByYear W30835440132020 @default.
• W3083544013 countsByYear W30835440132021 @default.
• W3083544013 crossrefType "posted-content" @default.

• next