[Research] Upcoming Lab meetings: March 3rd and March 10th

[Artur Dubrawski]

March 3rd, 12 noon, NSH 1507:

Speaker: Tzu-Kuo Huang
Auton Lab

Title: Learning Linear Dynamical Systems without Sequence Information

Abstract:
Virtually all methods of learning dynamic systems from data start from the
same basic assumption: that the learning algorithm will be provided with a
sequence, or trajectory, of data generated from the dynamic system.  In
this paper we consider the case where the data is not sequenced.  The
learning algorithm is presented a set of data points from the system's
operation but with no temporal ordering.  The data are simply drawn as
individual disconnected points.
While making this assumption may seem absurd at first glance, we observe
that many scientific modeling tasks have exactly this property.  In this
paper we restrict our attention to learning linear, discrete time models.
We propose several algorithms for learning these models based on optimizing
approximate likelihood functions and test the methods on several synthetic
data sets.

----

March 10th, 12 noon, NSH 1507:

Speaker: Marek Druzdzel
Associate Professor, Decision Systems Laboratory, School of Information Sciences, U. of Pittsburgh

Title: Cool Things That You Can Do With Graphical Probabilistic Models

Abstract:
In this introductory-level talk, I will give an overview of some of the 
research done in the last decade at the Decision Systems Laboratory, 
focusing on directed probabilistic graphs, such as Bayesian networks.  I 
will review basic knowledge engineering and inference techniques for 
Bayesian networks, influence diagrams, dynamic Bayesian networks, 
Bayesian networks involving equations and continuous distributions, 
along with applications in diagnosis, prognosis, data analysis, 
learning, and strategic planning.  I will show some user interface 
tricks for making the results of computation digestible for users.  As 
most of these topics are best explained "real-time", I will rely heavily 
on GeNIe, a graphical modeling environment developed at the Decision 
Systems Laboratory, and available at http://genie.sis.pitt.edu/.