RAMP: Boosting Adversarial Robustness Against Multiple
Enyi Jiang, Gagandeep Singh
We recommend first creating a conda environment using the provided environment.yml:
conda env create -f environment.yml
-
Main Result: The files
RAMP.pyandRAMP_wide_resnet.pyallow us to train ResNet-18 and WideReset models with standard choices of epsilons. To reproduce the results in the paper, one can runRAMP_scratch_cifar10.shin folderscripts/cifar10. -
Varying Epsilon Values: We provide scripts of
run_ramp_diff_eps_scratch.sh(RAMP),run_max_diff_eps_scratch.sh(MAX), andrun_eat_diff_eps_scratch.sh(E-AT) in folderscripts/cifar10for running the training from scratch experiments with different choices of epsilons.
-
To get pre-trained versions of ResNet-18 models with different epsilon values, one can run
pretrain_diff_eps_Lp.shscripts in folderscripts/cifar10. -
It is also possible to use models from the Model Zoo of RobustBench with
--model_name=RB_{}inserting the identifier of the classifier from the Model Zoo (these are automatically downloaded). (credits to E-AT paper) -
Main Result: To reproduce the results in the paper with different model architectures, one can run
RAMP_finetune_cifar10.shin folderscripts/cifar10andRAMP_finetune_imagenet.shin folderscripts/imagenet. -
Varying Epsilon Values: We provide scripts of
run_ramp_diff_eps_finetune.sh(RAMP),run_max_diff_eps_finetune.sh(MAX), andrun_eat_diff_eps_finetune.sh(E-AT) in folderscripts/cifar10for running the robust fine-tuning experiments with different choices of epsilons.
With --final_eval our standard evaluation (with APGD-CE and APGD-T, for a total of 10 restarts of 100 steps) is run for all threat models at the end of training.
Specifying --eval_freq=k a fast evaluation is run on test and training points every k epochs.
To evaluate a trained model one can run eval.py with --model_name as above for the pre-trained model or --model_name=/path/to/checkpoint/ for new or fine-tuned
classifiers. The corresponding architecture is loaded if the run has the automatically generated name. More details about the options for evaluation in eval.py.
Parts of the code in this repo is based on
Cite the paper/repo:
@article{jiang2024ramp,
title={RAMP: Boosting Adversarial Robustness Against Multiple $ l\_p $ Perturbations},
author={Jiang, Enyi and Singh, Gagandeep},
journal={arXiv preprint arXiv:2402.06827},
year={2024}
}