SNN对抗攻击笔记:
1. 解决SNN对抗攻击中脉冲与梯度数据格式不兼容性以及梯度消失问题:
2. 基于梯度的对抗攻击方式:
3. 图像转化到脉冲序列的采样方式:
4. SNN类型:
SNN-conv(ANN-SNN Conversion)[2-7, 9]
采用的神经元模型
Trick
SNN-BP[1-4, 8]
* * 分段常数函数(迭代LIF神经元模型)[1]
5. 影响对抗攻击效果的因素分析:
6.对抗攻击类型:
7. SNN/ANN-crafted攻击的比较:
8. SNN/ANN健壮性的比较:
9. SNN中神经元的选择:
10. SNN中池化的选择:
SNN
SNN-conv
SNN-BP
11. SNN中偏差的选择:
SNN
SNN-conv
SNN-BP
去除所有卷积层和全连接层中的偏差[7]
12. SNN中正则器的选择
SNN
SNN-conv
SNN-BP
Reference:
[1] Liang L , Hu X , Deng L , et al. Exploring Adversarial Attack in Spiking Neural Networks with Spike-Compatible Gradient[J]. 2020.
[2] Sharmin S , Rathi N , Panda P , et al. Inherent Adversarial Robustness of Deep Spiking Neural Networks: Effects of Discrete Input Encoding and Non-Linear Activations[J]. 2020.
[3] Sharmin S , Panda P , Sarwar S S , et al. A Comprehensive Analysis on Adversarial Robustness of Spiking Neural Networks[J]. 2019.
[4] Rathi, N., Srinivasan, G., Panda, P., Roy, K.: Enabling deep spiking neural networks with hybrid conversion and spike timing dependent backpropagation. In:ICLR (2020).
[5] Abhronil Sengupta, Yuting Ye, RobertWang, Chiao Liu, and Kaushik Roy. Going deeper in spiking neural networks: Vgg and residual architectures. Frontiers in neuroscience, 13, 2019.
[6] Peter U Diehl, Daniel Neil, Jonathan Binas, Matthew Cook, Shih-Chii Liu, and Michael Pfeiffer. Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing. In 2015 International Joint Conference on Neural Networks (IJCNN), pp. 1–8. IEEE, 2015.
[7] Cao Y , Chen Y , Khosla D . Spiking Deep Convolutional Neural Networks for Energy-Efficient Object Recognition[J]. International Journal of Computer Vision, 2015, 113(1):54-66.
[8] Chankyu Lee, Syed Shakib Sarwar, Kaushik Roy, Enabling Spike-based Backpropagation in State-of-the-art Deep Neural Network Architectures, arXiv:1903.06379, 2019.
[9] B. Rueckauer, I.-A. Lungu, Y. Hu, M. Pfeiffer, and S.-C. Liu. Conversion of continuous-valued deep networks to efficient event-driven networks for image classification. Frontiers in neuroscience, 11:682, 2017.
手机扫一扫
移动阅读更方便
你可能感兴趣的文章