[Colloquium] 10/11 Talks at TTIC: Brian Ziebart, University of Illinois at Chicago

[Mary Marre via Colloquium]

*When:    *  Thursday, October 11th at 11:00 am



*Where:     *TTIC, 6045 S. Kenwood Avenue, 5th Floor, Room 526



*Who:        *Brian Ziebart, University of Illinois at Chicago


*Title:* Supervised Structured Prediction Games

*Abstract:* Many important inductive reasoning problems, ranging from
understanding text and images, to enabling desirable robotic behavior, are
structured prediction tasks. Prevalent existing methods for supervised
structured prediction (e.g., conditional random fields, structured support
vector machines, and connectionist methods) learn by minimizing convex
surrogates of the application loss function—effectively optimizing over the
exact training data using an approximation of the loss. This talk will
present an approach based on the reverse: enabling structured prediction
for bipartite matchings, graph cuts, and graphical models by optimizing the
exact loss function using an adversarial approximation of the training data.

*Bio:* Brian Ziebart is an Associate Professor in the Department of
Computer Science at the University of Illinois at Chicago. He received his
PhD from Carnegie Mellon University where he was also a postdoctoral
fellow. He has published over 35 articles in leading machine learning and
artificial intelligence venues, including a 2011 Best Paper at the
International Conference on Machine Learning. This research is supported by
a 2017 NSF CAREER Award.

*Host:* Kevin Gimpel <kgimpel at ttic.edu>


Mary C. Marre
Administrative Assistant
*Toyota Technological Institute*
*6045 S. Kenwood Avenue*
*Room 517*
*Chicago, IL  60637*
*p:(773) 834-1757*
*f: (773) 357-6970*
*mmarre at ttic.edu <mmarre at ttic.edu>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20181005/cbac1633/attachment.html>