论文标题
在神经网络中设计特洛伊探测器的科学计算器
Scientific Calculator for Designing Trojan Detectors in Neural Networks
论文作者
论文摘要
这项工作提出了基于Web的交互式神经网络（NN）计算器和NN效率低下测量，该测量已被研究，目的是检测NN模型中嵌入的木马。该NN计算器的设计在Tensorflow Playground的顶部，具有数据和NN图的内存存储以及系数。它“像科学计算器”，具有分析，可视化和在训练数据集和NN体系结构上执行的输出操作。该原型在https://pages.nist.gov/nn-calculator上是可访问的。分析能力包括使用用于NN模型状态的直方图的修改后的kullback-liebler（KL）差异对NN效率低效率的新测量，以及对与数据和NN相关的变量的敏感性的量化。 NN计算器和KL差异均用于为各种特洛伊木马嵌入的特洛伊探测器方法设计。实验结果记录了KL差异测量相对于NN架构和数据集扰动的理想特性，以及有关嵌入式木马的推论。
This work presents a web-based interactive neural network (NN) calculator and a NN inefficiency measurement that has been investigated for the purpose of detecting trojans embedded in NN models. This NN Calculator is designed on top of TensorFlow Playground with in-memory storage of data and NN graphs plus coefficients. It is "like a scientific calculator" with analytical, visualization, and output operations performed on training datasets and NN architectures. The prototype is aaccessible at https://pages.nist.gov/nn-calculator. The analytical capabilities include a novel measurement of NN inefficiency using modified Kullback-Liebler (KL) divergence applied to histograms of NN model states, as well as a quantification of the sensitivity to variables related to data and NNs. Both NN Calculator and KL divergence are used to devise a trojan detector approach for a variety of trojan embeddings. Experimental results document desirable properties of the KL divergence measurement with respect to NN architectures and dataset perturbations, as well as inferences about embedded trojans.
