arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.05264cs.CRcs.AIcs.ET
收藏

安全感知近似脉冲神经网络

Security-Aware Approximate Spiking Neural Networks

University of Missouri(密苏里大学)

Syed Tihaam Ahmad, Ayesha Siddique, Khaza Anuarul Hoque

摘要:

深度神经网络(DNN)和脉冲神经网络(SNN)都因其对对抗攻击的敏感性而闻名。因此,最近的研究人员已经广泛地研究了DNN和SNN在对抗攻击下的鲁棒性和防御。与精确SNN(AccSNN)相比,近似SNN(AxSNN)已知对于超低功耗应用而言能效高达4倍。不幸的是,AxSNN在对抗攻击下的鲁棒性尚未被探索。在本文中,我们首先广泛地分析了AxSNNs的鲁棒性与不同的结构参数和近似水平下的两个基于梯度和两个神经形态攻击。然后,我们提出了两种新的防御方法,即,精确缩放和近似量化感知滤波(AQF),用于保护AxSNN。我们使用静态和神经形态数据集评估了这两种防御方法的有效性。我们的研究结果表明,AxSNNs更容易受到对抗攻击比AccSNNs,但精度缩放和AQF显着提高AxSNNs的鲁棒性。例如,与没有任何攻击的AccSNN相比,对AxSNN的PGD攻击导致72%的准确性损失,而对精确缩放的AxSNN的相同攻击仅导致静态MNIST数据集中17%的准确性损失(4倍鲁棒性改进)。类似地,当与没有任何攻击的AccSNN相比时,对AxSNN的稀疏攻击导致77%的准确性损失,而对具有AQF的AxSNN的相同攻击仅导致神经形态DVS 128手势数据集中的2%的准确性损失(38倍的鲁棒性改进)。

英文摘要:

Deep Neural Networks (DNNs) and Spiking Neural Networks (SNNs) are both known for their susceptibility to adversarial attacks. Therefore, researchers in the recent past have extensively studied the robustness and defense of DNNs and SNNs under adversarial attacks. Compared to accurate SNNs (AccSNN), approximate SNNs (AxSNNs) are known to be up to 4X more energy-efficient for ultra-low power applications. Unfortunately, the robustness of AxSNNs under adversarial attacks is yet unexplored. In this paper, we first extensively analyze the robustness of AxSNNs with different structural parameters and approximation levels under two gradient-based and two neuromorphic attacks. Then, we propose two novel defense methods, i.e., precision scaling and approximate quantization-aware filtering (AQF), for securing AxSNNs. We evaluated the effectiveness of these two defense methods using both static and neuromorphic datasets. Our results demonstrate that AxSNNs are more prone to adversarial attacks than AccSNNs, but precision scaling and AQF significantly improve the robustness of AxSNNs. For instance, a PGD attack on AxSNN results in a 72\% accuracy loss compared to AccSNN without any attack, whereas the same attack on the precision-scaled AxSNN leads to only a 17\% accuracy loss in the static MNIST dataset (4X robustness improvement). Similarly, a Sparse Attack on AxSNN leads to a 77\% accuracy loss when compared to AccSNN without any attack, whereas the same attack on an AxSNN with AQF leads to only a 2\% accuracy loss in the neuromorphic DVS128 Gesture dataset (38X robustness improvement).

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。

↑