Generalization to out-of-distribution (OOD) data is a capability natural to humans yet
challenging for machines to reproduce. This is because most learning algorithms strongly …

A Hilbert space embedding of a distribution—in short, a kernel mean embedding—has
recently emerged as a powerful tool for machine learning and statistical inference. The basic …

Machine learning systems generally assume that the training and testing distributions are
the same. To this end, a key requirement is to develop models that can generalize to unseen …

The goal of domain generalization algorithms is to predict well on distributions different from
those seen during training. While a myriad of domain generalization algorithms exist …

Standard training via empirical risk minimization (ERM) can produce models that achieve
low error on average but high error on minority groups, especially in the presence of …

Traditional machine learning paradigms are based on the assumption that both training and
test data follow the same statistical pattern, which is mathematically referred to as …

Though convolutional neural networks (CNNs) have demonstrated remarkable ability in
learning discriminative features, they often generalize poorly to unseen domains. Domain …

Distribution shifts—where the training distribution differs from the test distribution—can
substantially degrade the accuracy of machine learning (ML) systems deployed in the wild …

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 …

Neural network classifiers can largely rely on simple spurious features, such as
backgrounds, to make predictions. However, even in these cases, we show that they still …