# EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examples

Source: [EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examples](<https://www.sota2.com/research/paper/68fa64051e209131cea020b9>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T20:54:10.749Z

Authors: Pin-Yu Chen, Yash Sharma, Huan Zhang, Jinfeng Yi, Cho-Jui Hsieh

Published: 2017-09-13T02:40:59.000Z

## Abstract

Recent studies have highlighted the vulnerability of deep neural networks (DNNs) to adversarial examples - a visually indistinguishable adversarial image can easily be crafted to cause a well-trained model to misclassify. Existing methods for crafting adversarial examples are based on $L\_2$ and $L\_\\infty$ distortion metrics. However, despite the fact that $L\_1$ distortion accounts for the total variation and encourages sparsity in the perturbation, little has been developed for crafting $L\_1$-based adversarial examples. In this paper, we formulate the process of attacking DNNs via adversarial examples as an elastic-net regularized optimization problem. Our elastic-net attacks to DNNs (EAD) feature $L\_1$-oriented adversarial examples and include the state-of-the-art $L\_2$ attack as a special case. Experimental results on MNIST, CIFAR10 and ImageNet show that EAD can yield a distinct set of adversarial examples with small $L\_1$ distortion and attains similar attack performance to the state-of-the-art methods in different attack scenarios. More importantly, EAD leads to improved attack transferability and complements adversarial training for DNNs, suggesting novel insights on leveraging $L\_1$ distortion in adversarial machine learning and security implications of DNNs.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1709.04114>)

[PDF](<https://arxiv.org/pdf/1709.04114.pdf>)
