Advbox: a toolbox to generate adversarial examples that fool neural networks

TL;DR

Advbox is a comprehensive toolbox for generating adversarial examples across multiple deep learning frameworks, enabling robustness testing and attack scenario simulations including black box attacks and face recognition.

Contribution

It introduces a versatile, multi-framework toolbox supporting various attack scenarios, including black box attacks and face recognition, enhancing robustness evaluation of neural networks.

Findings

Supports multiple deep learning frameworks

Enables black box attack simulations

Includes diverse attack scenarios like DeepFake detection

Abstract

In recent years, neural networks have been extensively deployed for computer vision tasks, particularly visual classification problems, where new algorithms reported to achieve or even surpass the human performance. Recent studies have shown that they are all vulnerable to the attack of adversarial examples. Small and often imperceptible perturbations to the input images are sufficient to fool the most powerful neural networks. \emph{Advbox} is a toolbox to generate adversarial examples that fool neural networks in PaddlePaddle, PyTorch, Caffe2, MxNet, Keras, TensorFlow and it can benchmark the robustness of machine learning models. Compared to previous work, our platform supports black box attacks on Machine-Learning-as-a-service, as well as more attack scenarios, such as Face Recognition Attack, Stealth T-shirt, and DeepFake Face Detect. The code is licensed under the Apache 2.0 and is openly available at https://github.com/advboxes/AdvBox. Advbox now supports Python 3.