In this talk, I will present our recent research efforts of developing a deep learning library called LibAUC, which is applicable for solving a variety of compositional measures.  I will first present some background about optimization for machine learning and then introduce you a broad family of compositional measures/objectives called X-risk. Then I will focus on optimizing two measures, AUROC and AUPRC by talking about our theoretical research for designing algorithms and improving complexities, and our applied research for winning medical AI challenges.

Speaker Bio:

Tianbao Yang is an Associate Professor of Computer Science and Engineering at Texas A&M University. His research interests center round optimization, machine learning and AI. He received the best student paper award at COLT in 2012, NSF Career Award in 2019, and was named Dean’s Excellence in Research Scholar.  He is one of the early pioneers in Federated learning by publishing the first paper that inspires the Federated learning paradigm.  His group recently developed a deep learning library LibAUC, which has been downloaded more than 16K times and has won several competitions.  He has published more than 140 papers and served as associate editor of Neurocomputing,  and area chairs of ICML/NeurIPS/AAAI/IJCAI.

About Data Science Seminar

This seminar series will be run as weekly sessions and is hosted by ITEE Data Science.

Online via Zoom