迈向对抗鲁棒的持续学习
Towards Adversarially Robust Continual Learning
AI总结：
首次研究持续学习中的对抗鲁棒性问题，提出Task-Aware Boundary Augmentation (TABA)方法增强模型防御能力，并在CIFAR-10和CIFAR-100上验证了对抗训练与TABA的有效性。
摘要：
最近的研究表明，通过持续学习训练的模型能够取得与标准监督学习相当的性能，并且持续学习模型的学习灵活性使其在现实世界中获得了广泛应用。然而，深度学习模型已被证明容易受到对抗攻击的影响。尽管在标准监督学习背景下已有许多关于模型鲁棒性的研究，但保护持续学习免受对抗攻击的问题尚未得到探讨。为了填补这一研究空白，我们首次研究了持续学习中的对抗鲁棒性，并提出了一种名为Task-Aware Boundary Augmentation (TABA)的新方法来增强持续学习模型的鲁棒性。通过在CIFAR-10和CIFAR-100上进行的大量实验，我们展示了对抗训练和TABA在防御对抗攻击方面的有效性。
英文摘要：
Recent studies show that models trained by continual learning can achieve the comparable performances as the standard supervised learning and the learning flexibility of continual learning models enables their wide applications in the real world. Deep learning models, however, are shown to be vulnerable to adversarial attacks. Though there are many studies on the model robustness in the context of standard supervised learning, protecting continual learning from adversarial attacks has not yet been investigated. To fill in this research gap, we are the first to study adversarial robustness in continual learning and propose a novel method called \textbf{T}ask-\textbf{A}ware \textbf{B}oundary \textbf{A}ugmentation (TABA) to boost the robustness of continual learning models. With extensive experiments on CIFAR-10 and CIFAR-100, we show the efficacy of adversarial training and TABA in defending adversarial attacks.
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。
