对比预训练模型中的数据集所有权验证
Dataset Ownership Verification in Contrastive Pre-trained Models
AI总结：
针对现有方法无法验证无监督对比预训练模型数据集所有权的问题，提出基于嵌入空间实例关系差异的验证方法，在多模型上验证其p值显著低于0.05，性能优于现有方法，助力数据所有者维权。
摘要：
高质量开源数据集需要大量的整理工作，已成为深度学习快速发展的主要催化剂。同时，保护这些数据集对于数据所有者的利益至关重要。数据集所有权验证成为该领域的关键方法，但现有方法通常局限于监督模型，无法直接扩展到日益流行的无监督预训练模型。在这项工作中，我们提出了首个专为对比学习自监督预训练模型设计的数据集所有权验证方法。其主要目标是确定可疑的黑盒骨干网络是否在特定未标记数据集上预训练，以帮助数据所有者维护其权利。该方法的灵感来自我们的实证见解：当模型使用目标数据集训练时，嵌入空间中的一元和二元实例关系与未使用目标数据集训练的模型相比表现出显著差异。我们在多个对比预训练模型（包括SimCLR、BYOL、SimSiam、MOCO v3和DINO）上验证了该方法的有效性。结果表明，我们的方法以显著低于0.05的p值拒绝零假设，优于所有先前的方法。我们的代码可在https://github.com/xieyc99/DOV4CL获取。
英文摘要：
High-quality open-source datasets, which necessitate substantial efforts for curation, has become the primary catalyst for the swift progress of deep learning. Concurrently, protecting these datasets is paramount for the well-being of the data owner. Dataset ownership verification emerges as a crucial method in this domain, but existing approaches are often limited to supervised models and cannot be directly extended to increasingly popular unsupervised pre-trained models. In this work, we propose the first dataset ownership verification method tailored specifically for self-supervised pre-trained models by contrastive learning. Its primary objective is to ascertain whether a suspicious black-box backbone has been pre-trained on a specific unlabeled dataset, aiding dataset owners in upholding their rights. The proposed approach is motivated by our empirical insights that when models are trained with the target dataset, the unary and binary instance relationships within the embedding space exhibit significant variations compared to models trained without the target dataset. We validate the efficacy of this approach across multiple contrastive pre-trained models including SimCLR, BYOL, SimSiam, MOCO v3, and DINO. The results demonstrate that our method rejects the null hypothesis with a $p$-value markedly below $0.05$, surpassing all previous methodologies. Our code is available at https://github.com/xieyc99/DOV4CL.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
