Note: the 🌐 website provides a better UX experience for these docs.
These materials can be used either as a stand-alone and self-contained course on adversarial AI, or as a hands-on component of a more detailed university-level course on adversarial AI and related topics. This project began from student lecture notes from Dr. Allison Bishop’s graduate Adversarial AI course at the City College of New York.
For Instructors: See the 📚 Instructor Manual for accessing encrypted solutions and course management.
For Contributors: Check out our 🤝 Contributing Guide to help improve the course materials.
- 🐛 Issues: GitHub Issues
- Python 3 experience
- Basic understanding of data structures and algorithms
- Calculus 1 and Introductory Probability
Note: no prior experience with machine learning or deep learning is required, we want to make these materials as self contained as possible.
Location: part1/
Learn fundamental adversarial techniques through hash function vulnerabilities.
Components:
- 📚 Interactive recitation with live attack demonstrations
- 📝 Hands-on programming assignment
Location: part2/
A self-contained introduction to building a neural network.
Components:
- 📚 Interactive recitation building a convolutional neural network.
- 📝 Hands-on programming assignment
Location: part3/
Generate adversarial examples that fool deep learning models.
Components:
- 📚 Interactive recitation demonstrating adversarial attacks.
- 📝 Hands-on programming assignment
See AUTHORS.md for detailed credits.
These materials are for educational purposes only. The goal is for students to understand and defend against security vulnerabilities in machine learning.