# Meta Gradient Adversarial Attack

Source: [Meta Gradient Adversarial Attack](<https://www.sota2.com/research/paper/68fa6a941e209131cea29092>)

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

Data updated: 2025-10-25T23:28:26.804Z

Authors: Zheng Yuan, Jie Zhang, Yunpei Jia, Chuanqi Tan, Tao Xue, Shiguang Shan

Published: 2021-08-09T17:44:19.000Z

## Abstract

In recent years, research on adversarial attacks has become a hot spot. Although current literature on the transfer-based adversarial attack has achieved promising results for improving the transferability to unseen black-box models, it still leaves a long way to go. Inspired by the idea of meta-learning, this paper proposes a novel architecture called Meta Gradient Adversarial Attack (MGAA), which is plug-and-play and can be integrated with any existing gradient-based attack method for improving the cross-model transferability. Specifically, we randomly sample multiple models from a model zoo to compose different tasks and iteratively simulate a white-box attack and a black-box attack in each task. By narrowing the gap between the gradient directions in white-box and black-box attacks, the transferability of adversarial examples on the black-box setting can be improved. Extensive experiments on the CIFAR10 and ImageNet datasets show that our architecture outperforms the state-of-the-art methods for both black-box and white-box attack settings.

## 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/2108.04204>)

[PDF](<https://arxiv.org/pdf/2108.04204.pdf>)
