CRAICVLGSep 27, 2022

Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protection

Tsinghua
arXiv:2210.00875v3153 citationsh-index: 47Has Code
Originality Incremental advance
AI Analysis

This work addresses dataset copyright protection for data owners by introducing a stealthier and less harmful verification method, though it is incremental as it builds upon existing backdoor watermarking techniques.

The paper tackles the security risks of existing dataset copyright protection methods by proposing an untargeted backdoor watermarking scheme that avoids deterministic abnormal behaviors in deep neural networks, achieving effective ownership verification and resistance to backdoor defenses as verified on benchmark datasets.

Deep neural networks (DNNs) have demonstrated their superiority in practice. Arguably, the rapid development of DNNs is largely benefited from high-quality (open-sourced) datasets, based on which researchers and developers can easily evaluate and improve their learning methods. Since the data collection is usually time-consuming or even expensive, how to protect their copyrights is of great significance and worth further exploration. In this paper, we revisit dataset ownership verification. We find that existing verification methods introduced new security risks in DNNs trained on the protected dataset, due to the targeted nature of poison-only backdoor watermarks. To alleviate this problem, in this work, we explore the untargeted backdoor watermarking scheme, where the abnormal model behaviors are not deterministic. Specifically, we introduce two dispersibilities and prove their correlation, based on which we design the untargeted backdoor watermark under both poisoned-label and clean-label settings. We also discuss how to use the proposed untargeted backdoor watermark for dataset ownership verification. Experiments on benchmark datasets verify the effectiveness of our methods and their resistance to existing backdoor defenses. Our codes are available at \url{https://github.com/THUYimingLi/Untargeted_Backdoor_Watermark}.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes