论文标题
像素到二进制嵌入CNN的鲁棒性
Pixel to Binary Embedding Towards Robustness for CNNs
论文作者
论文摘要
卷积神经网络（CNN）的鲁棒性存在几个问题。例如，可以通过向输入中添加少量噪声来更改CNN的预测，当输入分布通过在训练过程中从未见过的转换移动时，CNN的性能会降低（例如，模糊效应）。有一些方法可以用二进制嵌入来替换像素值，以解决对抗性扰动的问题，从而成功改善了鲁棒性。在这项工作中，我们将像素提出到二进制嵌入（P2BE）以提高CNN的鲁棒性。 P2BE是一种可学习的二进制嵌入方法，而不是先前的手工编码的二进制嵌入方法。 P2BE在鲁棒性方面的表现优于其他二进制嵌入方法，以抵抗训练期间未显示的对抗性扰动和视觉损坏。
There are several problems with the robustness of Convolutional Neural Networks (CNNs). For example, the prediction of CNNs can be changed by adding a small magnitude of noise to an input, and the performances of CNNs are degraded when the distribution of input is shifted by a transformation never seen during training (e.g., the blur effect). There are approaches to replace pixel values with binary embeddings to tackle the problem of adversarial perturbations, which successfully improve robustness. In this work, we propose Pixel to Binary Embedding (P2BE) to improve the robustness of CNNs. P2BE is a learnable binary embedding method as opposed to previous hand-coded binary embedding methods. P2BE outperforms other binary embedding methods in robustness against adversarial perturbations and visual corruptions that are not shown during training.
