Removing Batch Normalization Boosts Adversarial Training

Authors: Haotao Wang, Aston Zhang, Shuai Zheng, Xingjian Shi, Mu Li, Zhangyang Wang

ICML 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that No Frost achieves a better accuracy-robustness trade-off compared with previous stateof-the-art AT methods based on BN or MBN models.
Researcher Affiliation Collaboration 1University of Texas at Austin, Austin, USA 2Amazon Web Services, Santa Clara, USA.
Pseudocode No The paper describes methods in text and equations but does not include structured pseudocode or algorithm blocks.
Open Source Code Yes Code and pretrained models are public1. 1https://github.com/amazon-research/normalizer-free-robust-training
Open Datasets Yes All methods are trained on the Image Net (Deng et al., 2009) dataset.
Dataset Splits Yes We evaluate clean accuracy on the Image Net validation set, and use the accuracy on adversarial test images as a metric for adversarial robustness.
Hardware Specification Yes All experiments are conducted with 8 NVIDIA V100 GPUs.
Software Dependencies No The paper mentions optimizers (SGD) but does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow, CUDA versions).
Experiment Setup Yes For all experiments, we train on Image Net for 90 epochs. We use the SGD optimizer with momentum 0.9. Batch size is 256. Weight decay factor is 5e-5. The initial learning rate is 0.1 and decays following a cosine annealing scheduler.