# GAP++: Learning to generate target-conditioned adversarial examples

arXiv: https://arxiv.org/abs/2006.05097 · HTML: https://mlsanity.com/paper/2006.05097/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Jun 2020.

## Built on by (most cited first, 5 of 5)

- [A Survey on Transferability of Adversarial Examples across Deep Neural Networks](https://mlsanity.com/paper/2310.17626/) (2023)
- [Comparative Evaluation of Recent Universal Adversarial Perturbations in Image Classification](https://mlsanity.com/paper/2306.11261/) (2023)
- [Benchmarking Adversarial Robustness of Compressed Deep Learning Models](https://mlsanity.com/paper/2308.08160/) (2023)
- [Sparse and Transferable Universal Singular Vectors Attack](https://mlsanity.com/paper/2401.14031/) (2024)
- [Towards Efficiently Evaluating the Robustness of Deep Neural Networks in IoT Systems: A GAN-based Method](https://mlsanity.com/paper/2111.10055/) (2021)

## Abstract (arXiv)

Adversarial examples are perturbed inputs which can cause a serious threat for machine learning models. Finding these perturbations is such a hard task that we can only use the iterative methods to traverse. For computational efficiency, recent works use adversarial generative networks to model the distribution of both the universal or image-dependent perturbations directly. However, these methods generate perturbations only rely on input images. In this work, we propose a more general-purpose framework which infers target-conditioned perturbations dependent on both input image and target label. Different from previous single-target attack models, our model can conduct target-conditioned attacks by learning the relations of attack target and the semantics in image. Using extensive experiments on the datasets of MNIST and CIFAR10, we show that our method achieves superior performance with single target attack models and obtains high fooling rates with small perturbation norms.