Skip to content Skip to navigation
Computer Science Department Colloquium
4/17/2014 04:10 pm
CBIM Multipurpose Room ( Room 22 )

Reliable Differential Dependency Network Analysis

Dr. Alexandru Niculescu-Mizil, NEC Research

Faculty Host: Tina Eliassi-Rad

Abstract

What is different between gene regulatory networks in cancer and healthy tissue? What is the difference between the functional brain network of people who suffer from schizophrenia and healthy people? I will talk about how these and other similar questions can be addressed via differential dependency network analysis, how naive approaches are prone to high false discovery rates that can render the analysis useless, and how differential dependency network analysis can be performed reliably by borrowing ideas and algorithms from multi-task learning.  I will show real case studies from oncology and neuroscience where the proposed technique have been used to extract differential networks that shed light on the biological processes involved in cancer and brain function.

This is joint work with Diane Owen, Rachel Ostroff, Alex Stewart and Vince P. Clark.

Bio

Alexandru Niculescu-Mizil is researcher at NEC Laboratories America. Before joining NEC, he was a Herman Goldstine postdoctoral fellow at IBM T.J. Watson Research Center. He received his Ph.D. from Cornell University in 2008 under the supervision of Rich Caruana, a Masters of Science degree in Computer Science from Cornell University and a Magna Cum Laude Bachelors degree in Mathematics and Computer Science from University of Bucharest. His research interests are in machine learning and data mining, particularly in inductive transfer, graphical model structure learning, probability estimation, empirical evaluations, ensemble methods and on-line learning. He received an ICML Distinguished Student Paper Award in 2005 for his work on probability estimation, and a COLT Best Student in 2008 paper award for his work on on-line learning. In 2009, he led the IBM Research team that won the KDDCUP competition.