Adversarial Robustness Toolbox (ART) is a Python library for Machine Learning Security. ART provides tools that enable developers and researchers to evaluate, defend, certify and verify 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, sci-kit-learn, XGBoost, LightGBM, CatBoost, GPy, etc.), all data types (images, tables, audio, video, etc.) and machine learning tasks (classification, object detection, generation, certification, etc.).
Features
- Adversarial Threats
- ART for Red and Blue Teams (selection)
- The library is under continuous development
- Documentation available
- Examples available
License
MIT LicenseFollow Adversarial Robustness Toolbox
Other Useful Business Software
Managed MySQL, PostgreSQL, and SQL Databases on Google Cloud
Cloud SQL is a fully managed relational database for MySQL, PostgreSQL, and SQL Server. We handle patching, backups, replication, encryption, and failover—so you can focus on your app. Migrate from on-prem or other clouds with free Database Migration Service. IDC found customers achieved 246% ROI. New customers get $300 in credits plus a 30-day free trial.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of Adversarial Robustness Toolbox!