Abstract Fair Principal Component Analysis (PCA) is a problem setting where we aim to
perform PCA while making the resulting representation fair in that the projected distributions …

We consider a semi-supervised classification problem with non-stationary label-shift in
which we observe a labelled data set followed by a sequence of unlabelled covariate …

Proper learning refers to the setting in which learners must emit predictors in the underlying
hypothesis class $ H $, and often leads to learners with simple algorithmic forms (eg …

High dimensional expanders (HDX) are a nascent generalization of expander graphs
(sparse yet robustly connected networks that play a core role in the theory of computation) to …

Recent success of machine learning is driven by scaling laws: larger architectures trained
using more data and compute lead to more “intelligent” agents. Therefore, even minor …