arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.10262cs.CVcs.AIcs.CRcs.LG
收藏

神经网络的零空间性质及其在图像隐写术中的应用

Null Space Properties of Neural Networks with Applications to Image Steganography

University of New Hampshire(新罕布什尔大学)

Xiang Li, Kevin M. Short

AI总结:

本文将零空间概念扩展至非线性神经网络,利用输入中不影响预测的零空间分量实现图像隐写,并在 MNIST 上证明网络可被诱导识别隐藏类别。

摘要:

本文探讨神经网络的零空间性质。我们将零空间定义从线性映射扩展到非线性映射,并讨论神经网络中零空间的存在性。给定神经网络的零空间能够揭示输入数据中对最终预测没有贡献的部分,因此我们可以利用它来欺骗神经网络。这揭示了神经网络中一种可被利用的固有弱点。本文描述的一个应用由此产生了一种图像隐写术方法。通过在 MNIST 等图像数据集上进行实验,我们表明可以使用零空间分量迫使神经网络选择一个预先选定的隐藏图像类别,即使整体图像可以被制作成看起来完全是另一幅图像。最后,我们比较了人类观看者会看到的内容,以及神经网络实际用于进行预测的图像部分,从而证明神经网络所“看见”的内容与我们预期的完全不同。

英文摘要:

This paper explores the null space properties of neural networks. We extend the null space definition from linear to nonlinear maps and discuss the presence of a null space in neural networks. The null space of a given neural network can tell us the part of the input data that makes no contribution to the final prediction so that we can use it to trick the neural network. This reveals an inherent weakness in neural networks that can be exploited. One application described here leads to a method of image steganography. Through experiments on image datasets such as MNIST, we show that we can use null space components to force the neural network to choose a selected hidden image class, even though the overall image can be made to look like a completely different image. We conclude by showing comparisons between what a human viewer would see, and the part of the image that the neural network is actually using to make predictions and, hence, show that what the neural network ``sees'' is completely different than what we would expect.

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。

↑