all AI news
Watch Out! Simple Horizontal Class Backdoors Can Trivially Evade Defenses
April 24, 2024, 4:43 a.m. | Hua Ma, Shang Wang, Yansong Gao, Zhi Zhang, Huming Qiu, Minhui Xue, Alsharif Abuadbba, Anmin Fu, Surya Nepal, Derek Abbott
cs.LG updates on arXiv.org arxiv.org
Abstract: All current backdoor attacks on deep learning (DL) models fall under the category of a vertical class backdoor (VCB) -- class-dependent. In VCB attacks, any sample from a class activates the implanted backdoor when the secret trigger is present. Existing defense strategies overwhelmingly focus on countering VCB attacks, especially those that are source-class-agnostic. This narrow focus neglects the potential threat of other simpler yet general backdoor types, leading to false security implications. This study introduces …
abstract arxiv attacks backdoor class cs.cr cs.lg current deep learning defense focus implanted sample secret simple strategies type
More from arxiv.org / cs.LG updates on arXiv.org
The Perception-Robustness Tradeoff in Deterministic Image Restoration
2 days, 22 hours ago |
arxiv.org
Jobs in AI, ML, Big Data
Founding AI Engineer, Agents
@ Occam AI | New York
AI Engineer Intern, Agents
@ Occam AI | US
AI Research Scientist
@ Vara | Berlin, Germany and Remote
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Machine Learning Engineer
@ Apple | Sunnyvale, California, United States