edcd3f3eac
* Pin the version of Tensorflow at >= 1.15. This allows us to remove a workaround that causes breakages for newer versions (we get errors if `AddV2` already has a registered gradient). * Pin the version of Tensorflow at < 2 since cleverhans <= 3.1.0 doesn't properly support TF2 * Pin the version of cleverhans between 3.0.1 and 3.1.0. Versions newer than 3.1.0 will no longer support TF1. * Comment out the dependency on `tensorflow-gpu` as it's not a strict requirement, and it does nothing for us on Travis. PiperOrigin-RevId: 368241595
Unsupervised Adversarial Training (UAT)
This repository contains the trained model and dataset used for Unsupervised Adversarial Training (UAT) from the paper Are Labels Required for Improving Adversarial Robustness?
Contents
This repo serves two primary functions:
- Data release: We share indices for the 80 Million Tiny Images Dataset subset used in our experiments, and a utility for loading the data.
- Model release: We have released our top-performing model on TF-Hub, and include an example demonstrating how to use it.
Running the code
Using the model
Our model is available via
TF-Hub.
For example usage, refer to quick_eval_cifar.py. The preferred method of
running this script is through run.sh, which will set up a virtual
environment, install the dependendencies, and run the evaluation script, which
will print the adversarial accuracy of the model.
cd /path/to/deepmind_research
unsupervised_adversarial_training/run.sh
Viewing the dataset
First, download the 80 Million Tiny Images Dataset image binary from the official web page: http://horatio.cs.nyu.edu/mit/tiny/data/index.html
Note this file is very large, and requires 227 GB of disc space.
The file tiny_200K_idxs.txt indicates which images from the dataset form the
80M@200K training set used in the paper. For example usage, refer to
save_example_images.py.
To view example images from this dataset, use the command:
cd /path/to/deepmind_research
python -m unsupervised_adversarial_training.save_example_images \
--data_bin_path=/path/to/tiny_images.bin
This will save the first 100 images to the directory
unsupervised_adversarial_training/images.
Citing this work
If you use this code in your work, please cite the accompanying paper:
@inproceedings{uat2019,
title={Are Labels Required for Improving Adversarial Robustness?},
author={Jonathan Uesato and Jean-Baptiste Alayrac and Po-Sen Huang and
Robert Stanforth and Alhussein Fawzi and Pushmeet Kohli},
booktitle={Advances in Neural Information Processing Systems},
year={2019}
}
Disclaimer
This is not an official Google product.