arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2411.09585cs.CR
收藏

基于距离驱动的解毒方法进行后门缓解

Backdoor Mitigation by Distance-Driven Detoxification

Shaokui Wei, Jiayin Liu, Hongyuan Zha

AI总结:

针对后门攻击破坏机器学习模型完整性的问题,提出距离驱动解毒(D3)方法,将防御重构为约束优化问题,实验表明其性能优于现有训练后防御技术。

摘要:

后门攻击通过使用中毒训练数据操纵预测,破坏机器学习模型的完整性。当特定触发器存在时,此类攻击导致目标误分类,而模型在其他条件下表现正常。本文考虑训练后后门防御任务,旨在清除预训练模型中的后门。我们首先分析普通微调的潜在问题,观察到它常陷入清洁样本和中毒样本损失均较低的区域。受此启发,我们提出距离驱动解毒(D3)方法,将后门防御重构为约束优化问题。具体而言,D3促使模型偏离初始权重附近,有效减少后门影响。在各种模型架构和数据集上针对最先进(SOTA)后门攻击的大量实验表明,D3不仅匹配且常优于现有SOTA训练后防御技术的性能。

英文摘要:

Backdoor attacks undermine the integrity of machine learning models by allowing attackers to manipulate predictions using poisoned training data. Such attacks lead to targeted misclassification when specific triggers are present, while the model behaves normally under other conditions. This paper considers a post-training backdoor defense task, aiming to detoxify the backdoors in pre-trained models. We begin by analyzing the underlying issues of vanilla fine-tuning and observe that it is often trapped in regions with low loss for both clean and poisoned samples. Motivated by such observations, we propose Distance-Driven Detoxification (D3), an innovative approach that reformulates backdoor defense as a constrained optimization problem. Specifically, D3 promotes the model's departure from the vicinity of its initial weights, effectively reducing the influence of backdoors. Extensive experiments on state-of-the-art (SOTA) backdoor attacks across various model architectures and datasets demonstrate that D3 not only matches but often surpasses the performance of existing SOTA post-training defense techniques.

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。

↑