arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.08016cs.CVcs.LGeess.IV
收藏

一种针对神经网络编码器的Out-of-Distribution攻击算法

An Algorithm for Out-Of-Distribution Attack to Neural Network Encoder

University of Miami(迈阿密大学)

Liang Liang, Linhai Ma, Linchen Qian, Jiasong Chen

摘要:

深度神经网络(DNN),特别是卷积神经网络,在图像分类任务上取得了优异的性能。然而,只有当训练模型的输入与训练样本相似时,才能保证这种性能,即,输入遵循训练集的概率分布。分布外(OOD)样本不遵循训练集的分布,因此OOD样本上的预测类标签变得毫无意义。已经提出了基于分类的方法用于OOD检测;然而,在这项研究中,我们表明这种类型的方法没有理论保证,并且由于DNN模型中的降维,我们的OOD攻击算法实际上是可破坏的。我们还表明,辉光似然为基础的OOD检测是易碎的。

英文摘要:

Deep neural networks (DNNs), especially convolutional neural networks, have achieved superior performance on image classification tasks. However, such performance is only guaranteed if the input to a trained model is similar to the training samples, i.e., the input follows the probability distribution of the training set. Out-Of-Distribution (OOD) samples do not follow the distribution of training set, and therefore the predicted class labels on OOD samples become meaningless. Classification-based methods have been proposed for OOD detection; however, in this study we show that this type of method has no theoretical guarantee and is practically breakable by our OOD Attack algorithm because of dimensionality reduction in the DNN models. We also show that Glow likelihood-based OOD detection is breakable as well.

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。

↑