EECS 281A / Stat 241A
      Statistical Learning Theory - Graphical Models      
Fall 2008
|
[Syllabus]
[Homework]
[Announcements]
[Projects]
|
Current and previous announcements:
- IMPORTANT: Sahand will not be holding office hours tomorrow (Wednesday 12/03/2008).
- Homework clarification. For HW 5 problem 1, we are designing an accept reject algorithm given a target distribution proportional to f(x)=x(1-x) and a proposed distribution g(x) = 1, where all values are taken over x \in [0,1]. We need to find an M such that f(x) \leq M g(x) \forall x. Of course, there are an infinite number of possible values that M can take. Hence, for part b), we select M such that if the proposed value is 1/2, the probability of accepting that value is 1/2.
- Please note that the Tuesday poster session time has changed to 1pm-3pm.
- You can sign up for the poster session at Poster Signups. It doesn't matter what number you sign up under, just what day. Thanks.
People
Professor:
Martin Wainwright
(wainwrig AT SYMBOL eecs DOT berkeley DOT edu)
Offices: 263 Cory Hall, 3-1978; 421 Evans Hall, 3-1975
Office hours: Tuesday, 11:00 - 12:00, 421 Evans Hall; Thursday, 15:30-16:30, Location: Start after class in 3 LeConte. Afterwards, check 330 Evans (for larger group) or 421 Evans (Prof. Wainwright's office).
TA:
Sahand Negahban (sahand_n AT SYMBOL eecs DOT berkeley DOT edu)
Office hours: Wednesday, 13:30-14:30 611 Soda Hall (Alcove)
TA:
Oleg Mayba
(oleg AT SYMBOL stat DOT berkeley DOT edu)
Office hours: Mondays, 10:00-11:00 387 Evans Hall.
Back to top
Practical information
Course description: This course is a 3-unit course that
provides an introduction to the area of probabilistic models based on
graphs. These graphical models provide a very flexible and powerful
framework for capturing statistical dependencies in complex,
multivariate data. Key issues to be addressed include representation,
efficient algorithms, inference and statistical estimation. These
concepts will be illustrated using examples drawn from various
application domains, including machine learning, signal processing,
communication theory, computational biology, computer vision etc.
Outline:
Basics on graphical models, Markov properties, recursive decomposability,
elimination algorithms
Sum-product algorithm, factor graphs, semi-rings
Markov properties of graphical models
Junction tree algorithm
Chains, trees, factorial models, coupled models, layered models
Kalman filtering and Rauch-Tung-Striebel smoothing
Hidden Markov models (HMM) and forward-backward
Exponential family, sufficiency, conjugacy
Frequentist and Bayesian methods
The EM algorithm
Conditional mixture models, hierarchical mixture models
Factor analysis, principal component analysis (PCA), canonical correlation analysis (CCA), independent component analysis (ICA)
Importance sampling, Gibbs sampling, Metropolis-Hastings
Variational algorithms: mean field, belief propagation, convex relaxations
Dynamical graphical models
Model selection, marginal likelihood, AIC, BIC and MDL
Applications to signal processing, bioinformatics,
communication, computer vision etc.
Volume: 3 units
Lectures: 3 LeConte, Tues, Thurs 14:00-15:30.
Section: Wednesday, 17:00-18:00, 330 Evans Hall.
Grading: Homework and Course Project.
Required background:
The prerequisites are previous coursework in linear algebra,
multivariate calculus, and basic probability and statistics. Previous
coursework in graph theory, information theory, optimization theory
and statistical physics would be helpful but is not required.
Familiarity with Matlab, Splus or a related matrix-oriented
programming language will be necessary.
Back to top
Last modified: Wed Sept, 3rd