/adversarial-robustness-toolbox

Python library for adversarial attacks and defenses (evasion, poisoning) for neural networks with multiple framework support

Primary LanguagePythonMIT LicenseMIT

Adversarial Robustness Toolbox (ART) v1.7


Continuous Integration CodeQL Documentation Status PyPI Language grade: Python Total alerts codecov Code style: black License: MIT PyPI - Python Version slack-img Downloads Downloads

中文README请按此处

Adversarial Robustness Toolbox (ART) is a Python library for Machine Learning Security. ART provides tools that enable developers and researchers to defend and evaluate Machine Learning models and applications against the adversarial threats of Evasion, Poisoning, Extraction, and Inference. ART supports all popular machine learning frameworks (TensorFlow, Keras, PyTorch, MXNet, scikit-learn, XGBoost, LightGBM, CatBoost, GPy, etc.), all data types (images, tables, audio, video, etc.) and machine learning tasks (classification, object detection, speech recognition, generation, certification, etc.).

Adversarial Threats


ART for Red and Blue Teams (selection)


Learn more

Get Started Documentation Contributing
- Installation
- Examples
- Notebooks		 - Attacks
- Defences
- Estimators
- Metrics
- Technical Documentation		 - Slack, Invitation
- Contributing
- Roadmap
- Citing

The library is under continuous development. Feedback, bug reports and contributions are very welcome!

Acknowledgment

This material is partially based upon work supported by the Defense Advanced Research Projects Agency (DARPA) under Contract No. HR001120C0013. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Defense Advanced Research Projects Agency (DARPA).