San Francisco
Chapter of the American Statistical Association September Seminar
|
Speaker: |
Jie Peng, Department of Statistics,
University of California, Davis |
|
Time: |
Monday
September 29, 2014, 4:30-5:00 pm refreshments and networking, 5:00-6:00pm talk |
|
Location: |
Stanford
HRP (Health Research & Policy) Conference room T138B map |
|
Title: |
Learning graphical
models |
Abstract:
A fundamental question in Statistics is
to understand how a set of random variables are
interacting with each other. Interactions (in terms of conditional
dependencies) can be represented by graphs which provide compact
representations of a probability distribution. Graphical models have wide applications
in many fields including genomics, neuroscience, artificial intelligence, etc.
In this talk, we will start with an
introduction of graphical models, focusing on Markov random fields and Bayesian
networks. We will discuss the connection between these two types of models
and various algorithms for fitting them. Particularly, we will emphasize
structure learning algorithms and the high-dimensional regime. Numerical
examples and applications to genomics data will be mentioned throughout
the talk.