Data poisoning is an adversarial scenario where an attacker feeds a specially crafted sequence of samples to an online model in order to subvert learning. We introduce Lethean Attack, a novel data poisoning technique that induces catastrophic forgetting on an online model. We apply the attack in the context of Test-Time Training, a modern online learning framework aimed for generalization under distribution shifts. We present the theoretical rationale and empirically compare it against other sample sequences that naturally induce forgetting. Our results demonstrate that using lethean attacks, an adversary could revert a test-time training model back to coin-flip accuracy performance using a short sample sequence.
翻译:数据中毒是一种对抗性假设,攻击者将专门制作的样本序列输入在线模型,以颠覆学习。我们引入了Lethean Apection,这是一种新颖的数据中毒技术,在网上模型中诱发灾难性的遗忘。我们在测试-时间培训中应用了这一攻击,这是一个现代在线学习框架,目的是在分布变换中进行概括化。我们展示了理论原理,并从经验上将其与其他自然诱发遗忘的样本序列进行比较。我们的结果表明,使用 Lethean攻击,对手可以将测试-时间培训模型恢复到使用简短样本序列的硬翻精确性能。