Publication
Algorithms
Publication
Sparse PSD approximation of the PSD cone
Project
Machine Learning Theory
Our work in machine learning theory ranges from modeling to mathematical analysis of algorithms to experimental probing. The diverse set of topics include clustering, optimization, unsupervised learning, algebraic methods, fairness, causality, and deep learning.
Project
Extreme Classification
xtreme classification is a rapidly growing research area in computer vision focusing on multi-class and multi-label problems involving an extremely large number of labels (ranging from thousands to billions).