yamamura-k/ACG

PyTorch implementation of Diversified Adversarial Attack based on Conjugate Gradient Method (ICML2022).

36
/ 100
Emerging

This project helps machine learning engineers and researchers assess the robustness of their image classification models against sophisticated adversarial attacks. It takes an image dataset (like ImageNet, CIFAR-10, or CIFAR-100) and a pre-trained image classification model as input. The output is a quantitative measure of the model's vulnerability, specifically the attack success rates of diversified adversarial attacks.

No commits in the last 6 months.

Use this if you need to rigorously evaluate how well your deep learning image classification models withstand advanced, 'black-box' adversarial attacks, and identify potential weaknesses.

Not ideal if you are looking for a tool to defend against adversarial attacks, or for general model interpretability, rather than solely evaluating attack success.

AI-model-evaluation adversarial-robustness computer-vision deep-learning-security image-classification
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 15 / 25

How are scores calculated?

Stars

10

Forks

4

Language

Python

License

MIT

Last pushed

Jun 26, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/yamamura-k/ACG"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.