The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization

Abstract

We introduce four new real-world distribution shift datasets consisting of changes in image style, image blurriness, geographic location, camera operation, and more. With our new datasets, we take stock of previously proposed methods for improving out-of-distribution robustness and put them to the test. We find that using larger models and artificial data augmentations can improve robustness on real-world distribution shifts, contrary to claims in prior work. We find improvements in artificial robustness benchmarks can transfer to real-world distribution shifts, contrary to claims in prior work. Motivated by our observation that data augmentations can help with real-world distribution shifts, we also introduce a new data augmentation method which advances the state-of-the-art and outperforms models pretrained with 1000 times more labeled data. Overall we find that some methods consistently help with distribution shifts in texture and local image statistics, but these methods do not help with some other distribution shifts like geographic changes. Our results show that future research must study multiple distribution shifts simultaneously, as we demonstrate that no evaluated method consistently improves robustness.

ICCV 2021; Datasets, code, and models available at https://github.com/hendrycks/imagenet-r

Cited by in corpus (13)

• RobustBench: a standardized adversarial robustness benchmark
• Partial success in closing the gap between human and machine vision
• RobustART: Benchmarking Robustness on Architecture Design and Training Techniques
• Understanding and Improving Robustness of Vision Transformers through Patch-based Negative Augmentation
• Structure discovery in Atomic Force Microscopy imaging of ice
• Discrete Representations Strengthen Vision Transformer Robustness
• Certified Adversarial Defenses Meet Out-of-Distribution Corruptions: Benchmarking Robustness and Simple Baselines
• Training on Test Data with Bayesian Adaptation for Covariate Shift
• A Fine-Grained Analysis on Distribution Shift
• Understanding and Testing Generalization of Deep Networks on Out-of-Distribution Data
• Pyramid Adversarial Training Improves ViT Performance
• Characterizing and Improving the Robustness of Self-Supervised Learning through Background Augmentations
• Covariate Shift in High-Dimensional Random Feature Regression