2021
DOI: 10.48550/arxiv.2109.00685
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Excess Capacity and Backdoor Poisoning

Abstract: A backdoor data poisoning attack is an adversarial attack wherein the attacker injects several watermarked, mislabeled training examples into a training set. The watermark does not impact the test-time performance of the model on typical data; however, the model reliably errs on watermarked examples.To gain a better foundational understanding of backdoor data poisoning attacks, we present a formal theoretical framework within which one can discuss backdoor data poisoning attacks for classification problems. We… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...

Citation Types

0
0
0

Publication Types

Select...

Relationship

0
0

Authors

Journals

citations
Cited by 0 publications
references
References 14 publications
0
0
0
Order By: Relevance

No citations

Set email alert for when this publication receives citations?