maestro-student

This repository contains the student version of the maestro platform to learn adversarial machine learning algorithms. The paper "Maestro: a gamified platform for teaching AI robustness" (link) was accepted at the Thirteenth Symposium on Educational Advances in Artificial Intelligence, EAAI 2023.

Installation

To install the Maestro Platform, follow these steps:

1. Create a python environment by conda or virtualenv.

# take conda as an example
conda create -n maestro-student python==3.9.11
conda activate maestro-student

2. Clone the repository:

git clone [email protected]:C0ldstudy/maestro-student.git

3. Navigate to the project directory

4. Install the required python packages

pip install -r requirements.txt

Usage

We list six tasks for students to complete. It also supports adding other assigments follow our basic code structure. For each task, we provide a brief README file that includes the background information, required datasets, and evaluation requirements. To gain a deeper understanding of each task, we also list related knowledge and papers for further reading.

• Attack Homework: Gentic Algorithm
• Attack Project: PGD, CW.
• Defense Homework/Project: Adversarial Training

Teacher Version

If you are interested in obtaining the teacher version, please send a request to [email protected].

Citation

@inproceedings{DBLP:conf/aaai/GeletaXLW00M23,
  author       = {Margarita Geleta and
                  Jiacen Xu and
                  Manikanta Loya and
                  Junlin Wang and
                  Sameer Singh and
                  Zhou Li and
                  Sergio Gago Masagu{\'{e}}},
  editor       = {Brian Williams and
                  Yiling Chen and
                  Jennifer Neville},
  title        = {Maestro: {A} Gamified Platform for Teaching {AI} Robustness},
  booktitle    = {Thirty-Seventh {AAAI} Conference on Artificial Intelligence, {AAAI}
                  2023, Thirty-Fifth Conference on Innovative Applications of Artificial
                  Intelligence, {IAAI} 2023, Thirteenth Symposium on Educational Advances
                  in Artificial Intelligence, {EAAI} 2023, Washington, DC, USA, February
                  7-14, 2023},
  pages        = {15816--15824},
  publisher    = {{AAAI} Press},
  year         = {2023},
  url          = {https://doi.org/10.1609/aaai.v37i13.26878},
  doi          = {10.1609/AAAI.V37I13.26878},
  timestamp    = {Sun, 12 Nov 2023 02:11:30 +0100},
  biburl       = {https://dblp.org/rec/conf/aaai/GeletaXLW00M23.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}