判别式自监督学习在视觉中的对抗鲁棒性
Adversarial Robustness of Discriminative Self-Supervised Learning in Vision
AI总结：
本研究评估了七个判别式自监督模型与一个监督模型在多种视觉任务上的对抗鲁棒性，发现自监督模型在ImageNet分类和线性迁移学习中更具鲁棒性，但微调后差距缩小，分割检测中优势减弱。
摘要：
自监督学习（SSL）在视觉表示学习方面取得了显著进展，然而对其对抗鲁棒性的全面评估仍然有限。在本研究中，我们评估了七个判别式自监督模型和一个监督模型在多种任务上的对抗鲁棒性，包括ImageNet分类、迁移学习、分割和检测。我们的研究结果表明，在ImageNet上，判别式自监督模型通常比其监督对应模型表现出更好的对抗攻击鲁棒性，且这一优势在使用线性评估时扩展到迁移学习。然而，当应用微调时，自监督与监督模型之间的鲁棒性差距显著缩小。同样，这一鲁棒性优势在分割和检测任务中减弱。我们还研究了各种可能影响对抗鲁棒性的因素，包括架构选择、训练时长、数据增强和批量大小。我们的分析有助于持续探索视觉自监督表示系统中的对抗鲁棒性。
英文摘要：
Self-supervised learning (SSL) has advanced significantly in visual representation learning, yet comprehensive evaluations of its adversarial robustness remain limited. In this study, we evaluate the adversarial robustness of seven discriminative self-supervised models and one supervised model across diverse tasks, including ImageNet classification, transfer learning, segmentation, and detection. Our findings suggest that discriminative SSL models generally exhibit better robustness to adversarial attacks compared to their supervised counterpart on ImageNet, with this advantage extending to transfer learning when using linear evaluation. However, when fine-tuning is applied, the robustness gap between SSL and supervised models narrows considerably. Similarly, this robustness advantage diminishes in segmentation and detection tasks. We also investigate how various factors might influence adversarial robustness, including architectural choices, training duration, data augmentations, and batch sizes. Our analysis contributes to the ongoing exploration of adversarial robustness in visual self-supervised representation systems.
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。
