Artificial Intelligence (AI) is making a profound impact in almost every domain. A vital enabler
of its great success is the availability of abundant and high-quality data for building machine …

The two fields of machine learning and graphical causality arose and are developed
separately. However, there is, now, cross-pollination and increasing interest in both fields to …

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 …

We present a simple self-training method that achieves 88.4% top-1 accuracy on ImageNet,
which is 2.0% better than the state-of-the-art model that requires 3.5 B weakly labeled …

Data augmentation is a simple yet effective way to improve the robustness of deep neural
networks (DNNs). Diversity and hardness are two complementary dimensions of data …

We evaluate a wide range of ImageNet models with five trained human labelers. In our year-
long experiment, trained humans first annotated 40,000 images from the ImageNet and …

We extend semi-supervised learning to the problem of domain adaptation to learn
significantly higher-accuracy models that train on one data distribution and test on a different …

Deploying machine learning systems in the real world requires both high accuracy on clean
data and robustness to naturally occurring corruptions. While architectural advances have …

Although machine learning models typically experience a drop in performance on out-of-
distribution data, accuracies on in-versus out-of-distribution data are widely observed to …

Traditional normalization techniques (eg, Batch Normalization and Instance Normalization)
generally and simplistically assume that training and test data follow the same distribution …