arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2012.10485cs.LGcs.AI
收藏

RAILS:一个鲁棒的对抗性免疫学习系统

RAILS: A Robust Adversarial Immune-inspired Learning System

University of Michigan(密歇根大学)

Ren Wang, Tianqi Chen, Stephen Lindsly, Alnawaz Rehemtulla, Alfred Hero, Indika Rajapakse

摘要:

针对深度神经网络的对抗攻击正在不断发展。如果没有有效的防御,它们可能导致灾难性的失败。哺乳动物的免疫系统是一种长期存在的、可以说是最强大的自然防御系统,数百万年来,它成功地抵御了新病原体的攻击。在本文中,我们提出了一个新的对抗防御框架,称为鲁棒对抗免疫启发学习系统(RAILS)。RAILS结合了自适应免疫系统仿真(AISE),它模拟了用于保护宿主免受病原体攻击的生物机制。我们使用RAILS来加固深度k最近邻(DkNN)架构以抵御规避攻击。进化编程用于模拟自然免疫系统中的过程:B细胞群集,克隆扩张和亲和力成熟。我们发现,RAILS学习曲线表现出类似的多样性选择学习阶段,在我们的体外生物实验中观察到的。当应用于三个不同数据集上的对抗性图像分类时,与单独应用DkNN相比,RAILS提供了额外的5.62%/12.56%/4.74%的鲁棒性改进,而对干净数据的准确性没有明显的损失。

英文摘要:

Adversarial attacks against deep neural networks are continuously evolving. Without effective defenses, they can lead to catastrophic failure. The long-standing and arguably most powerful natural defense system is the mammalian immune system, which has successfully defended against attacks by novel pathogens for millions of years. In this paper, we propose a new adversarial defense framework, called the Robust Adversarial Immune-inspired Learning System (RAILS). RAILS incorporates an Adaptive Immune System Emulation (AISE), which emulates in silico the biological mechanisms that are used to defend the host against attacks by pathogens. We use RAILS to harden Deep k-Nearest Neighbor (DkNN) architectures against evasion attacks. Evolutionary programming is used to simulate processes in the natural immune system: B-cell flocking, clonal expansion, and affinity maturation. We show that the RAILS learning curve exhibits similar diversity-selection learning phases as observed in our in vitro biological experiments. When applied to adversarial image classification on three different datasets, RAILS delivers an additional 5.62%/12.56%/4.74% robustness improvement as compared to applying DkNN alone, without appreciable loss of accuracy on clean data.

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。

↑