Pixel to Binary Embedding Towards Robustness for CNNs

TL;DR

This paper introduces Pixel to Binary Embedding (P2BE), a learnable binary embedding method that enhances CNN robustness against adversarial attacks and unseen visual corruptions, outperforming previous methods.

Contribution

The paper proposes P2BE, a novel learnable binary embedding approach that improves CNN robustness compared to prior hand-coded binary embeddings.

Findings

P2BE outperforms existing binary embeddings in robustness.

P2BE enhances CNN resilience against adversarial perturbations.

P2BE maintains performance under unseen visual corruptions.

Abstract

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.