Second-order adversarial attack and certifiable robustness

B Li; C Chen; W Wang; L Carin

Second-order adversarial attack and certifiable robustness

B Li, C Chen, W Wang, L Carin

Research output: Contribution to journal › Article › peer-review

Abstract

We propose a powerful second-order attack method that outperforms existing attack methods on reducing the accuracy of state-of-the-art defense models based on adversarial training. The effectiveness of our attack method motivates an investigation of provable …

Original language	Undefined/Unknown
Journal	arXiv preprint arXiv:1809.03113
State	Published - 2018
Externally published	Yes

Bibliographical note

Query date: 2021-03-11 11:12:31

Access to Document

https://www.researchgate.net/profile/Bai_Li11/publication/327570654_Second-Order_Adversarial_Attack_and_Certifiable_Robustness/links/5ba97e1392851ca9ed237217/Second-Order-Adversarial-Attack-and-Certifiable-Robustness.pdf

Cite this

@article{17d6e075ef8c4ef989b10cd3586480a5,

title = "Second-order adversarial attack and certifiable robustness",

abstract = "We propose a powerful second-order attack method that outperforms existing attack methods on reducing the accuracy of state-of-the-art defense models based on adversarial training. The effectiveness of our attack method motivates an investigation of provable …",

author = "B Li and C Chen and W Wang and L Carin",

note = "Query date: 2021-03-11 11:12:31",

year = "2018",

language = "Undefined/Unknown",

journal = "arXiv preprint arXiv:1809.03113",

publisher = "researchgate.net",

}

TY - JOUR

T1 - Second-order adversarial attack and certifiable robustness

AU - Li, B

AU - Chen, C

AU - Wang, W

AU - Carin, L

N1 - Query date: 2021-03-11 11:12:31

PY - 2018

Y1 - 2018

N2 - We propose a powerful second-order attack method that outperforms existing attack methods on reducing the accuracy of state-of-the-art defense models based on adversarial training. The effectiveness of our attack method motivates an investigation of provable …

AB - We propose a powerful second-order attack method that outperforms existing attack methods on reducing the accuracy of state-of-the-art defense models based on adversarial training. The effectiveness of our attack method motivates an investigation of provable …

M3 - Article

JO - arXiv preprint arXiv:1809.03113

JF - arXiv preprint arXiv:1809.03113

ER -