Adversarial Robustness 360


The open source Adversarial Robustness Toolbox provides tools that enable developers and researchers to evaluate and defend machine learning models and applications against the adversarial threats of evasion, poisoning, extraction, and inference.

 

Not sure what to do first? Start here!

Read More

Learn about adversarial robustness concepts, terminology, and tools before you begin.


*

Try Web Demos

Explore interactive web demos that illustrate the capabilities available in this toolbox.


*

Watch Videos

Watch videos to learn more about the Adversarial Robustness Toolbox.


*

Read a Paper

Read a paper describing the details of the Adversarial Robustness Toolbox.


*

Use Tutorials

Step through a set of in-depth examples to evaluate and defend AI models.


*

Ask a Question

Visit our Adversarial Robustness Toolbox Slack to ask questions, make comments, and tell stories about how you use the toolbox.


*

View Notebooks

Open a directory of Jupyter notebooks in GitHub that provide working examples of adversarial robustness. Then share your own notebooks!


*

Contribute

You can add new algorithms and metrics in GitHub. Share Jupyter notebooks showcasing how you have tested vulnerabilities of your machine learning application.


*

About this site

The Adversarial Robustness Toolbox (ART) was created by IBM Research and donated by IBM to the Linux Foundation AI & Data.

Additional research sites that advance other aspects of Trusted AI include:

AI Fairness 360
AI Privacy 360
AI Explainability 360
Uncertainty Quantification 360
AI FactSheets 360