Machine Learning

Statistical Guarantees for Local Graph Clustering

Parallel and Communication Avoiding Least Angle Regression

A Flexible Coordinate Descent Method

Avoiding Synchronization in First-Order Methods for Sparse Convex Optimization

Avoiding communication in primal and dual block coordinate descent methods

Performance of First- and Second-Order Methods for L1-Regularized Least Squares Problems

A Second-Order Method for Strongly-Convex L1-Regularization Problems