Manipulating Out-Domain Uncertainty Estimation in Deep Neural Networks via Targeted Clean-Label Poisoning-Reference-Cited by-同舟云学术

Manipulating Out-Domain Uncertainty Estimation in Deep Neural Networks via Targeted Clean-Label Poisoning

Published:2023-10-21 Issue: Volume: Page:
ISSN:
Container-title:Proceedings of the 32nd ACM International Conference on Information and Knowledge Management
language:
Short-container-title:

Author:

Zeng Huimin¹^ORCID,Yue Zhenrui¹^ORCID,Zhang Yang¹^ORCID,Shang Lanyu¹^ORCID,Wang Dong¹^ORCID

Affiliation:

1. University of Illinois Urbana-Champaign, Urbana-Champaign, IL, USA

Funder

NSF (National Science Foundation)

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3583780.3614957

Reference38 articles.

1. Ismail Alarab and Simant Prakoonwit . 2021. Adversarial Attack for Uncertainty Estimation: Identifying Critical Regions in Neural Networks. arXiv preprint arXiv:2107.07618 ( 2021 ). Ismail Alarab and Simant Prakoonwit. 2021. Adversarial Attack for Uncertainty Estimation: Identifying Critical Regions in Neural Networks. arXiv preprint arXiv:2107.07618 (2021).

2. Adversarial Robustness on In- and Out-Distribution Improves Explainability

3. Battista Biggio , Blaine Nelson , and Pavel Laskov . 2012. Poisoning attacks against support vector machines. arXiv preprint arXiv:1206.6389 ( 2012 ). Battista Biggio, Blaine Nelson, and Pavel Laskov. 2012. Poisoning attacks against support vector machines. arXiv preprint arXiv:1206.6389 (2012).

4. Certifiably adversarially robust detection of out-of-distribution data;Bitterwolf Julian;Advances in Neural Information Processing Systems,2020

5. John Bradshaw , Alexander G de G Matthews, and Zoubin Ghahramani . 2017 . Adversarial examples, uncertainty, and transfer testing robustness in gaussian process hybrid deep networks. arXiv preprint arXiv:1707.02476 (2017). John Bradshaw, Alexander G de G Matthews, and Zoubin Ghahramani. 2017. Adversarial examples, uncertainty, and transfer testing robustness in gaussian process hybrid deep networks. arXiv preprint arXiv:1707.02476 (2017).