/Adversarial-examples

Introduction to adversarial examples - Deep learning

Primary LanguageJupyter Notebook

Adversarial-examples

👨‍🎓 This project was carried out during my master's degree in computer vision at URJC - Madrid

Introduction to adversarial examples - Deepl learning

Goals

  • Changing the prediction result of an image classification without changing its visual to the human eye

Requirements

  • Python 3.7+
  • jupyter ~= 1.0.0
  • matplotlib ~= 3.3
  • tensorflow ~= 2.6

How to install all the requirements :

pip3 install -r requirements.txt

Usage

jupyter notebook Adversarial\ examples\ -\ Deep\ learning.ipynb

Results

Changing the result of the prediction to a random class (Non-directed), class 3 to 5 (random)

Changing the result of the prediction to a specific class (Directed), class 3 to 9

Structure

.
├── Adversarial examples - Deep learning.ipynb
├── imgs
│    ├── result1.png
│    └── result.png
├── README.md
└── requirements.txt

Authors