利用非完美恢复实现数据可用性攻击
Leveraging Imperfect Restoration for Data Availability Attack
浏览论文内容
中文总结 AI 辅助
针对现有数据可用性攻击CUDA在自监督学习场景效果差、图像质量与投毒效果存在权衡的问题，提出非完美恢复投毒（IRP）方法，经对比验证其在监督与自监督学习场景中均优于8种基线方法及5种防御方法。
中文摘要 AI 辅助
在线数据的丰富性面临着被未经授权用于训练深度学习模型的风险。为应对这一问题，各类数据可用性攻击（Data Availability Attacks, DAAs）已被设计出来，通过对训练数据进行细微扰动，使此类模型无法学习这些数据。然而，现有攻击通常在监督学习（Supervised Learning, SL）或自监督学习（Self-Supervised Learning, SSL）场景中表现出色，其中一种生成基于卷积的不可学习数据集（Convolution-based Unlearnable Dataset, CUDA）的无模型方法，是在SSL和SL场景中最具鲁棒性的DAA。尽管如此，CUDA对SSL的效果仍不理想，且在图像质量与投毒效果之间存在严重权衡。本文对CUDA进行了理论分析，揭示了其引入的次优梯度，并阐明了它为实现数据投毒所采用的逐类偏差策略。基于此，我们提出了一种名为非完美恢复投毒（Imperfect Restoration Poisoning, IRP）的新型投毒方法，旨在在保持高图像质量的同时实现强投毒效果。通过在SL和SSL场景中将IRP与8种基线方法进行广泛对比，并结合5种代表性防御方法的评估，我们展示了IRP的优越性。代码：this https URL
英文摘要
The abundance of online data is at risk of unauthorized usage in training deep learning models. To counter this, various Data Availability Attacks (DAAs) have been devised to make data unlearnable for such models by subtly perturbing the training data. However, existing attacks often excel against either Supervised Learning (SL) or Self-Supervised Learning (SSL) scenarios. Among these, a model-free approach that generates a Convolution-based Unlearnable Dataset (CUDA) stands out as the most robust DAA across both SSL and SL. Nonetheless, CUDA's effectiveness against SSL is underwhelming and it faces a severe trade-off between image quality and its poisoning effect. In this paper, we conduct a theoretical analysis of CUDA, uncovering the sub-optimal gradients it introduces and elucidating the strategy it employs to induce class-wise bias for data poisoning. Building on this, we propose a novel poisoning method named Imperfect Restoration Poisoning (IRP), aiming to preserve high image quality while achieving strong poisoning effects. Through extensive comparisons of IRP with eight baselines across SL and SSL, coupled with evaluations alongside five representative defense methods, we showcase the superiority of IRP. Code: https://github.com/lyumingzhi/IRP
发表机构
- College of Computing and Data Science, Nanyang Technological University（南洋理工大学计算与数据科学学院）
- Nanyang Technological University（南洋理工大学）
机构由 AI 辅助整理，请以论文原文为准。
