November 4, 2016, Arindam Banerjee, University of Minnesota

Friday, November 4, 2016, 2-3pm
8 St. Mary’s Street, PHO 404/428
Refreshments at 1:45pm

Arindam Banerjee
University of Minnesota

Learning with Low Samples in High-Dimensions: Estimators, Geometry, and Applications 

Many machine learning problems, especially scientific problems arising in areas such as climate science, ecology, brain sciences, and genomics, operate in the so-called `low sample, high dimension’ regime. Such problems typically have numerous possible predictors or features but the number of training examples is small, often much smaller than the number of features. In this talk, we will discuss recent advances in designing and analyzing estimators for such problems, which generalize prior work such as the Lasso and the Dantzig selector. We will discuss the geometry underlying such problems, and how the geometry helps in establishing sample complexity and finite sample estimation error bounds. We will also discuss applications of such results in structure learning in graphical models, multivariate time series models, and multi-task learning.
This is joint work with Sheng Chen, Farideh Fazayeli, Andre Goncalves, Igor Melnyk, and Vidyashankar Sivakumar.

Arindam Banerjee is an Associate Professor at the Department of Computer & Engineering and a Resident Fellow at the Institute on the Environment at the University of Minnesota, Twin Cities. His research interests are in statistical machine learning and data mining, and applications in complex real-world problems including climate sciences, ecology, recommendation systems, text analysis, brain sciences, finance, and aviation safety. He has won several awards, including the Adobe Research Award (2016), the IBM Faculty Award (2013), the NSF CAREER award (2010), and six Best Paper awards in top-tier conferences.

Faculty Host: Brian Kulis
Student Host: Andrew Cutler