Empirical risk minimization (ERM) is known in practice to be non-robust ...
Many modern machine learning tasks require models with high tail perform...
Many machine learning tasks involve subpopulation shift where the testin...
Many machine learning applications involve learning representations that...
Adversarial robustness has become a fundamental requirement in modern ma...
We address imbalanced classification, the problem in which a label may h...
Simulated annealing is an effective and general means of optimization. I...
Adversarial training is one of the most popular ways to learn robust mod...
We study the low rank approximation problem of any given matrix A over
R...
We develop a framework for learning sparse nonparametric directed acycli...
Neural network robustness has recently been highlighted by the existence...
We study the notion of Bilu-Linial stability in the context of Independe...
We study the sample complexity of semi-supervised learning (SSL) and
int...
Motivated by problems in data clustering, we establish general condition...