all InfoSec news
Backdoor Watermarking Deep Learning Classification Models With Deep Fidelity. (arXiv:2208.00563v1 [cs.CR])
Aug. 2, 2022, 1:20 a.m. | Guang Hua, Andrew Beng Jin Teoh
cs.CR updates on arXiv.org arxiv.org
Backdoor Watermarking is a promising paradigm to protect the copyright of
deep neural network (DNN) models for classification tasks. In the existing
works on this subject, researchers have intensively focused on watermarking
robustness, while fidelity, which is concerned with the original functionality,
has received less attention. In this paper, we show that the existing shared
notion of the sole measurement of learning accuracy is insufficient to
characterize backdoor fidelity. Meanwhile, we show that the analogous concept
of embedding distortion in …
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
Information Security Engineer (Vienna) - (m/f/d)
@ Sportradar | Wien, Poland
DevSecOps Engineer - U.S. Citizenship Required
@ Ardent MC | Remote
Head of AML, Regulatory and Compliance
@ Delivery Hero | Athens, Greece
Cybersecurity professional Mid-Senior level
@ Ethics Code | El Salvador - Remote
Senior Information Security Specialist
@ TRISTAR | 1801 Liberty Drive, Bloomington, IN, USA
SOC Analyst Level 2
@ Inbox Business Technologies | Islamabad, Islamabad Capital Territory, Pakistan