Hyperparticle/one-pixel-attack-keras
Keras implementation of "One pixel attack for fooling deep neural networks" using differential evolution on Cifar10 and ImageNet
This project explores how a deep neural network can be fooled into misclassifying an image by changing just one pixel. It takes an image and a target misclassification, then outputs a slightly altered image that causes the network to make the wrong prediction. This is for machine learning researchers and security analysts evaluating the robustness of image classification models.
1,232 stars. No commits in the last 6 months.
Use this if you need to understand the vulnerability of deep neural networks to minimal adversarial perturbations in image classification tasks.
Not ideal if you are looking for a general-purpose image augmentation or data anonymization tool.
Stars
1,232
Forks
215
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Apr 24, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Hyperparticle/one-pixel-attack-keras"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
Trusted-AI/adversarial-robustness-toolbox
Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion,...
bethgelab/foolbox
A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX
DSE-MSU/DeepRobust
A pytorch adversarial library for attack and defense methods on images and graphs
cleverhans-lab/cleverhans
An adversarial example library for constructing attacks, building defenses, and benchmarking both
BorealisAI/advertorch
A Toolbox for Adversarial Robustness Research