# Robustness of Rotation-Equivariant Networks to Adversarial Perturbations

arXiv: https://arxiv.org/abs/1802.06627 · HTML: https://mlsanity.com/paper/1802.06627/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Feb 2018.

## Built on by (most cited first, 6 of 6)

- [SynthID-Image: Image watermarking at internet scale](https://mlsanity.com/paper/2510.09263/) (2025)
- [E(2) Equivariant Neural Networks for Robust Galaxy Morphology Classification](https://mlsanity.com/paper/2311.01500/) (2023)
- [Position: Certified Robustness Does Not (Yet) Imply Model Security](https://mlsanity.com/paper/2506.13024/) (2025)
- [Is Certifying ℓp Robustness Still Worthwhile?](https://mlsanity.com/paper/2310.09361/) (2023)
- [SELTO: Sample-Efficient Learned Topology Optimization](https://mlsanity.com/paper/2209.05098/) (2022)
- [Lipschitz Optimization for Formal Verification of Homographies](https://mlsanity.com/paper/2605.23203/) (2026)

## Abstract (arXiv)

Deep neural networks have been shown to be vulnerable to adversarial examples: very small perturbations of the input having a dramatic impact on the predictions. A wealth of adversarial attacks and distance metrics to quantify the similarity between natural and adversarial images have been proposed, recently enlarging the scope of adversarial examples with geometric transformations beyond pixel-wise attacks. In this context, we investigate the robustness to adversarial attacks of new Convolutional Neural Network architectures providing equivariance to rotations. We found that rotation-equivariant networks are significantly less vulnerable to geometric-based attacks than regular networks on the MNIST, CIFAR-10, and ImageNet datasets.