Measuring Robustness to Natural Distribution Shifts in Image Classification

Rohan Taori, Achal Dave, Vaishaal Shankar, Nicholas Carlini, Benjamin Recht, Ludwig Schmidt

Keywords: machine learning, robustness, distribution shift, image corruptions, reliable machine learning

TL;DR: We compare many current robustness strategies and find that most do not reliably promote robustness to natural distribution shifts.

Abstract: We study how robust current ImageNet models are to distribution shifts arising from natural variations in datasets. Most research on robustness focuses on synthetic image perturbations (noise, simulated weather artifacts, adversarial examples, etc.), which leaves open how robustness on synthetic distribution shift relates to distribution shift arising in real data. Informed by an evaluation of 204 ImageNet models in 213 different test conditions, we find that there is often little to no transfer of robustness from current synthetic to natural distribution shift. Moreover, most current techniques provide no robustness to the natural distribution shifts in our testbed. The main exception is training on larger and more diverse datasets, which in multiple cases increases robustness, but is still far from closing the performance gaps. Our results indicate that distribution shifts arising in real data are currently an open research problem.

Submission Track: Published work track

Reviewer Emails: [e-mail]

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2007.00644/code)