NCSA staff who would like to submit an item for the calendar can email newsdesk@ncsa.illinois.edu.
Chris Fletcher
Arindam BanerjeeAbstract:In spite of the dramatic success of deep learning over the past decade, there is limitedunderstanding of why things are working. The talk will shed some light on two facets of the mystery:optimization and generalization. For optimization, we will discuss the low-dimensional geometryof gradients in deep learning, active vs. lazy learning, and some implications. For generalization,we discuss how smoothed analysis, which avoids worst-case analysis by adding a bit of noiseto problem components, may be an effective approach to understanding generalization in deepnetworks. We will discuss preliminary empirical and theoretical results on both the facets, anddiscuss directions for future work.
Bio:Arindam Banerjee is a Founder Professor at the Department of Computer Science, Universityof Illinois Urbana-Champaign. His research interests are in machine learning and data mining,especially on problems involving geometry and randomness. His current research focuses oncomputational and statistical aspects of deep learning, spatial and temporal data analysis, andsequential decision making problems. His work also focuses on applications in complex real-worldproblems in different areas including climate science, ecology, recommendation systems, andfinance, among others. He has won several awards, including the NSF CAREER award (2010), theIBM Faculty Award (2013), and six best paper awards in top-tier venues.