https://github.com/mishig25/adversarial-reprogramming-keras

Keras implementation of Adversarial Reprogramming of Neural Networks
https://github.com/mishig25/adversarial-reprogramming-keras

deep-learning deep-neural-networks

Last synced: 18 days ago
JSON representation

Keras implementation of Adversarial Reprogramming of Neural Networks

• Host: GitHub
• URL: https://github.com/mishig25/adversarial-reprogramming-keras
• Owner: mishig25
• Created: 2018-08-12T19:32:22.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-09-01T18:36:45.000Z (over 6 years ago)
• Last Synced: 2024-12-03T15:48:43.239Z (29 days ago)
• Topics: deep-learning, deep-neural-networks
• Language: Jupyter Notebook
• Size: 3.91 KB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Adversarial Reprogramming of Neural Networks

Keras implementation of Adversarial Reprogramming of Neural Networks.

Link to the paper:  [https://arxiv.org/abs/1806.11146](https://arxiv.org/abs/1806.11146)



Adversarial reporgramming demonstration. Source: arXiv:1806.11146

### Contents:

- `model.py` contains adv reprogramming model and [custom keras layer](https://keras.io/layers/writing-your-own-keras-layers/) for adv reprogramming

- `demo_nb.ipynb` is a Jupyter notebook for demonstrating contents and usage of `model.py`

### Prerequisites:

- Python 3.6

- Keras 2.2.0

- Tensorflow 1.8.0

### Tested on:

- OS: ubuntu 16.04 LTS

- GPU: single GeForce GTX 1070 

### References

- [Post by Rajat Dwaraknath](https://rajatvd.github.io/Exploring-Adversarial-Reprogramming/)

- [TF implementation by Abhishek Sinha](https://github.com/a7b23/adversarial-reprogramming)