Adversarial Reprogramming of Neural Networks

This is a PyTorch re-implementation for the ICLR'19 paper: Adversarial Reprogramming of Neural Networks provided by : kcelia 👩🏻‍💻 & VieVie31 👨🏻‍💻

Example of learned program from random noise on MNIST with SqueezeNet

Visualisation of the learning process

Poster of our re-implementation & experiments

@inproceedings{
    elsayed2018adversarial,
    title={Adversarial Reprogramming of Neural Networks},
    author={Gamaleldin F. Elsayed and Ian Goodfellow and Jascha Sohl-Dickstein},
    booktitle={International Conference on Learning Representations},
    year={2019},
    url={https://openreview.net/forum?id=Syx_Ss05tm},
}