A Bias-Variance Trade-Off View on Robustness Enhancement

This repo contain the code for the captioned paper.

Usage

Perform fast adversarial training on CIFAR10 with 3 splits of training data.

python train_at_fast.py n_split=3

Name		Name	Last commit message	Last commit date
Latest commit History 13 Commits
configs		configs
models		models
README.md		README.md
attack.py		attack.py
eval.py		eval.py
train_at_fast.py		train_at_fast.py
train_base.py		train_base.py
utils.py		utils.py