DeepEclipse: How to Break White-Box DNN-Watermarking Schemes

arXiv:2403.03590v1 Announce Type: new
Abstract: Deep Learning (DL) models have become crucial in digital transformation, thus raising concerns about their intellectual property rights. Different watermarking techniques have been developed to protect Deep Neural Networks (DNNs) from IP infringement, creating a competitive field for DNN watermarking and removal methods. The predominant watermarking schemes use white-box techniques, which involve modifying weights by adding a unique signature to specific DNN layers. On the other hand, existing attacks on white-box watermarking usually require knowledge …

arxiv box competitive cs.cr cs.lg deep learning digital digital transformation intellectual property networks neural networks property protect rights techniques transformation watermarking