通过弱监督对比对抗训练，提升半监督数据中鲁棒特征学习效果
Weakly Supervised Contrastive Adversarial Training for Learning Robust Features from Semi-supervised Data
- 基于信息论设计弱监督对比对抗训练，实现非鲁棒特征的完整扰动
- 在多个基准上验证了对鲁棒特征学习的显著提升
- 适合关注半监督学习与对抗鲁棒性提升的研究者
现有对抗训练方法常因扰动不完全，未能充分打乱非鲁棒特征与标签间的关联，导致鲁棒特征学习不充分。由于难以区分鲁棒与非鲁棒特征且标注数据稀疏，实现完全扰动极具挑战。为此，本文提出弱监督对比对抗训练（WSCAT），通过在部分标注数据上生成完整的对抗样本，破坏非鲁棒特征与标签间的残余相关性，从而实现更优的鲁棒特征学习。理论分析与广泛实验验证了WSCAT的有效性。代码已开源。
原文摘要 · Abstract (English)
Existing adversarial training (AT) methods often suffer from incomplete perturbation, meaning that not all non-robust features are perturbed when generating adversarial examples (AEs). This results in residual correlations between non-robust features and labels, leading to suboptimal learning of robust features. However, achieving complete perturbation, i.e., perturbing as many non-robust features as possible, is challenging due to the difficulty in distinguishing robust and non-robust features and the sparsity of labeled data. To address these challenges, we propose a novel approach called Weakly Supervised Contrastive Adversarial Training (WSCAT). WSCAT ensures complete perturbation for improved learning of robust features by disrupting correlations between non-robust features and labels through complete AE generation over partially labeled data, grounded in information theory. Extensive theoretical analysis and comprehensive experiments on widely adopted benchmarks validate the superiority of WSCAT. Our code is available at https://github.com/zhang-lilin/WSCAT.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。