Robust Perception through Equivariance

Authors: Chengzhi Mao, Lingyu Zhang, Abhishek Vaibhav Joshi, Junfeng Yang, Hao Wang, Carl Vondrick

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our empirical experiments show that restoring feature equivariance at inference time defends against worst-case adversarial perturbations. The method obtains improved adversarial robustness on four datasets (Image Net, Cityscapes, PASCAL VOC, and MS-COCO) on image recognition, semantic segmentation, and instance segmentation tasks.
Researcher Affiliation Academia 1Department of Computer Science, Columbia University, New York, USA 2Department of Computer Science, Rutgers University, New Jersey, USA.
Pseudocode Yes Algorithm 1 Equivariance Defense
Open Source Code Yes Our code is available at https: //github.com/cvlab-columbia/Equi4Rob.
Open Datasets Yes Our experiments evaluate the adversarial robustness on four datasets: Image Net (Deng et al., 2009), Cityscapes (Cordts et al., 2016), PASCAL-VOC (Everingham et al., 2010), and MS-COCO (Lin et al., 2014).
Dataset Splits No The paper mentions 'randomly sample 2% of data for evaluation' for ImageNet, but does not provide specific train/validation/test dataset splits (percentages or counts) for any of the datasets.
Hardware Specification Yes We evaluate on a single A6000 GPU.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes We choose the number of transformations to be K = 8, which empirically can be fit into a 2080Ti GPU with batch size 1. ... We use steps T = 20 for all our defense tasks.