English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

For machine learning systems to be reliable, we must understand theirperformance in unseen, out-of-distribution environments. In this paper, weempirically show that out-of-distribution performance is strongly correlatedwith in-distribution performance for a wide range of models and distributionshifts. Specifically, we demonstrate strong correlations betweenin-distribution and out-of-distribution performance on variants of CIFAR-10 &ImageNet, a synthetic pose estimation task derived from YCB objects, satelliteimagery classification in FMoW-WILDS, and wildlife classification iniWildCam-WILDS. The strong correlations hold across model architectures,hyperparameters, training set size, and training duration, and are more precisethan what is expected from existing domain adaptation theory. To complete thepicture, we also investigate cases where the correlation is weaker, forinstance some synthetic distribution shifts from CIFAR-10-C and the tissueclassification dataset Camelyon17-WILDS. Finally, we provide a candidate theorybased on a Gaussian data model that shows how changes in the data covariancearising from distribution shift can affect the observed correlations.

Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization