# Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNets

arXiv: https://arxiv.org/abs/2002.05990 · HTML: https://mlsanity.com/paper/2002.05990/

## Key facts

- Cited by 167 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Feb 2020.

## Built on by (most cited first, 30 of 167)

- [Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks](https://mlsanity.com/paper/2101.05930/) (2021)
- [Unlearnable Examples: Making Personal Data Unexploitable](https://mlsanity.com/paper/2101.04898/) (2021)
- [Admix: Enhancing the Transferability of Adversarial Attacks](https://mlsanity.com/paper/2102.00436/) (2021)
- [Frequency Domain Model Augmentation for Adversarial Attack](https://mlsanity.com/paper/2207.05382/) (2022)
- [Advances in adversarial attacks and defenses in computer vision: A survey](https://mlsanity.com/paper/2108.00401/) (2021)
- [Towards Transferable Adversarial Attacks on Vision Transformers](https://mlsanity.com/paper/2109.04176/) (2021)
- [RobustART: Benchmarking Robustness on Architecture Design and Training Techniques](https://mlsanity.com/paper/2109.05211/) (2021)
- [Improving Adversarial Robustness via Channel-wise Activation Suppressing](https://mlsanity.com/paper/2103.08307/) (2021)
- [On Improving Adversarial Transferability of Vision Transformers](https://mlsanity.com/paper/2106.04169/) (2021)
- [Boosting the Transferability of Adversarial Attacks with Reverse Adversarial Perturbation](https://mlsanity.com/paper/2210.05968/) (2022)
- [VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models](https://mlsanity.com/paper/2310.04655/) (2023)
- [Fight Back Against Jailbreaking via Prompt Adversarial Tuning](https://mlsanity.com/paper/2402.06255/) (2024)
- [On Generating Transferable Targeted Perturbations](https://mlsanity.com/paper/2103.14641/) (2021)
- [Structure Invariant Transformation for better Adversarial Transferability](https://mlsanity.com/paper/2309.14700/) (2023)
- [Adversarial Robustness Comparison of Vision Transformer and MLP-Mixer to CNNs](https://mlsanity.com/paper/2110.02797/) (2021)
- [Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bits](https://mlsanity.com/paper/2102.10496/) (2021)
- [Boosting Adversarial Transferability by Achieving Flat Local Maxima](https://mlsanity.com/paper/2306.05225/) (2023)
- [Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models](https://mlsanity.com/paper/2112.05588/) (2021)
- [A Survey on Transferability of Adversarial Examples across Deep Neural Networks](https://mlsanity.com/paper/2310.17626/) (2023)
- [Beyond ImageNet Attack: Towards Crafting Adversarial Examples for Black-box Domains](https://mlsanity.com/paper/2201.11528/) (2022)
- [Enhancing Adversarial Training with Second-Order Statistics of Weights](https://mlsanity.com/paper/2203.06020/) (2022)
- [Transferable Adversarial Attacks on Vision Transformers with Token Gradient Regularization](https://mlsanity.com/paper/2303.15754/) (2023)
- [LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity](https://mlsanity.com/paper/2207.13129/) (2022)
- [SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation](https://mlsanity.com/paper/2312.04913/) (2023)
- [Towards Understanding and Boosting Adversarial Transferability from a Distribution Perspective](https://mlsanity.com/paper/2210.04213/) (2022)
- [Transferable Adversarial Attack based on Integrated Gradients](https://mlsanity.com/paper/2205.13152/) (2022)
- [A Little Robustness Goes a Long Way: Leveraging Robust Features for Targeted Transfer Attacks](https://mlsanity.com/paper/2106.02105/) (2021)
- [DISCO: Adversarial Defense with Local Implicit Functions](https://mlsanity.com/paper/2212.05630/) (2022)
- [Diversifying the High-level Features for better Adversarial Transferability](https://mlsanity.com/paper/2304.10136/) (2023)
- [Analysis and Applications of Class-wise Robustness in Adversarial Training](https://mlsanity.com/paper/2105.14240/) (2021)

## Abstract (arXiv)

Skip connections are an essential component of current state-of-the-art deep neural networks (DNNs) such as ResNet, WideResNet, DenseNet, and ResNeXt. Despite their huge success in building deeper and more powerful DNNs, we identify a surprising security weakness of skip connections in this paper. Use of skip connections allows easier generation of highly transferable adversarial examples. Specifically, in ResNet-like (with skip connections) neural networks, gradients can backpropagate through either skip connections or residual modules. We find that using more gradients from the skip connections rather than the residual modules according to a decay factor, allows one to craft adversarial examples with high transferability. Our method is termed Skip Gradient Method(SGM). We conduct comprehensive transfer attacks against state-of-the-art DNNs including ResNets, DenseNets, Inceptions, Inception-ResNet, Squeeze-and-Excitation Network (SENet) and robustly trained DNNs. We show that employing SGM on the gradient flow can greatly improve the transferability of crafted attacks in almost all cases. Furthermore, SGM can be easily combined with existing black-box attack techniques, and obtain high improvements over state-of-the-art transferability methods. Our findings not only motivate new research into the architectural vulnerability of DNNs, but also open up further challenges for the design of secure DNN architectures.