FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2943281030> ?p ?o ?g. }

• W2943281030 abstract "A large body of recent work has investigated the phenomenon of evasion attacks using adversarial examples for deep learning systems, where the addition of norm-bounded perturbations to the test inputs leads to incorrect output classification. Previous work has investigated this phenomenon in closed-world systems where training and test inputs follow a pre-specified distribution. However, real-world implementations of deep learning applications, such as autonomous driving and content classification are likely to operate in the open-world environment. In this paper, we demonstrate the success of open-world evasion attacks, where adversarial examples are generated from out-of-distribution inputs (OOD adversarial examples). In our study, we use 11 state-of-the-art neural network models trained on 3 image datasets of varying complexity. We first demonstrate that state-of-the-art detectors for out-of-distribution data are not robust against OOD adversarial examples. We then consider 5 known defenses for adversarial examples, including state-of-the-art robust training methods, and show that against these defenses, OOD adversarial examples can achieve up to 4$times$ higher target success rates compared to adversarial examples generated from in-distribution data. We also take a quantitative look at how open-world evasion attacks may affect real-world systems. Finally, we present the first steps towards a robust open-world machine learning system." @default.
• W2943281030 created "2019-05-09" @default.
• W2943281030 creator A5003032696 @default.
• W2943281030 creator A5011437254 @default.
• W2943281030 creator A5015619835 @default.
• W2943281030 creator A5040461332 @default.
• W2943281030 creator A5052709754 @default.
• W2943281030 creator A5074489729 @default.
• W2943281030 creator A5089243170 @default.
• W2943281030 date "2019-05-05" @default.
• W2943281030 modified "2023-09-26" @default.
• W2943281030 title "Better the Devil you Know: An Analysis of Evasion Attacks using Out-of-Distribution Adversarial Examples." @default.
• W2943281030 cites W1503398984 @default.
• W2943281030 cites W1522301498 @default.
• W2943281030 cites W1551760018 @default.
• W2943281030 cites W1686810756 @default.
• W2943281030 cites W1968411139 @default.
• W2943281030 cites W2007657822 @default.
• W2943281030 cites W2091432990 @default.
• W2943281030 cites W2095577883 @default.
• W2943281030 cites W2101890615 @default.
• W2943281030 cites W2108598243 @default.
• W2943281030 cites W2116400298 @default.
• W2943281030 cites W2123045220 @default.
• W2943281030 cites W2158899491 @default.
• W2943281030 cites W2160815625 @default.
• W2943281030 cites W2163605009 @default.
• W2943281030 cites W2180612164 @default.
• W2943281030 cites W2194775991 @default.
• W2943281030 cites W2243397390 @default.
• W2943281030 cites W2274565976 @default.
• W2943281030 cites W2401231614 @default.
• W2943281030 cites W2408141691 @default.
• W2943281030 cites W2460937040 @default.
• W2943281030 cites W2511730936 @default.
• W2943281030 cites W2535873859 @default.
• W2943281030 cites W2543927648 @default.
• W2943281030 cites W2552767274 @default.
• W2943281030 cites W2557283755 @default.
• W2943281030 cites W2561498661 @default.
• W2943281030 cites W2565186948 @default.
• W2943281030 cites W2590523583 @default.
• W2943281030 cites W2593871270 @default.
• W2943281030 cites W2611110439 @default.
• W2943281030 cites W2612445135 @default.
• W2943281030 cites W2616841723 @default.
• W2943281030 cites W2618169590 @default.
• W2943281030 cites W2625220439 @default.
• W2943281030 cites W2736899637 @default.
• W2943281030 cites W2736941579 @default.
• W2943281030 cites W2738841453 @default.
• W2943281030 cites W2744095836 @default.
• W2943281030 cites W2747237166 @default.
• W2943281030 cites W2750384547 @default.
• W2943281030 cites W2765233338 @default.
• W2943281030 cites W2765384636 @default.
• W2943281030 cites W2766447205 @default.
• W2943281030 cites W2766462876 @default.
• W2943281030 cites W2767075075 @default.
• W2943281030 cites W2768899812 @default.
• W2943281030 cites W2773381986 @default.
• W2943281030 cites W2775467454 @default.
• W2943281030 cites W2783113218 @default.
• W2943281030 cites W2783882201 @default.
• W2943281030 cites W2784500886 @default.
• W2943281030 cites W2786088545 @default.
• W2943281030 cites W2786104118 @default.
• W2943281030 cites W2786163515 @default.
• W2943281030 cites W2786712888 @default.
• W2943281030 cites W2787496614 @default.
• W2943281030 cites W2787708942 @default.
• W2943281030 cites W2789159078 @default.
• W2943281030 cites W2791953061 @default.
• W2943281030 cites W2793931959 @default.
• W2943281030 cites W2794609696 @default.
• W2943281030 cites W2798356176 @default.
• W2943281030 cites W2798801120 @default.
• W2943281030 cites W2799107510 @default.
• W2943281030 cites W2803156329 @default.
• W2943281030 cites W2803697594 @default.
• W2943281030 cites W2805807555 @default.
• W2943281030 cites W2806075129 @default.
• W2943281030 cites W2867167548 @default.
• W2943281030 cites W2884821828 @default.
• W2943281030 cites W2886462939 @default.
• W2943281030 cites W2888940765 @default.
• W2943281030 cites W2890505418 @default.
• W2943281030 cites W2890884881 @default.
• W2943281030 cites W2895097814 @default.
• W2943281030 cites W2898963688 @default.
• W2943281030 cites W2899692219 @default.
• W2943281030 cites W2899771611 @default.
• W2943281030 cites W2903105043 @default.
• W2943281030 cites W2904981516 @default.
• W2943281030 cites W2910068345 @default.
• W2943281030 cites W2949103145 @default.
• W2943281030 cites W2949308370 @default.
• W2943281030 cites W2949311987 @default.
• W2943281030 cites W2949506549 @default.
• W2943281030 cites W2950029838 @default.

• next