|
Instructor:
Dan Gildea office hours Wed 1-2pm
Time: M/W 2:00-3:15, CSB 601
Motto: Meliora
Required text: Boyd and Vanderberghe, Convex Optimization
In addition to this text, we will cover additional numerical
methods taken from Nocedal and Wright and Numerical Recipes. The
second half of the course will focus on applications of
optimization in the context of machine learning.
| W 9/3 | Intro | Boyd ch 1
| M 9/8 | Convex Sets | Boyd ch 2
| W 9/10 | Convex Functions | Boyd ch 3
| M 9/15 | Convex Optimization Problems | Boyd ch 4
| W 9/17 | Convex Optimization Problems | Boyd ch 4
| M 9/22 | Duality | Boyd ch 5Tag |
| W 9/24 | Approximation and Statistical Estimation | Boyd ch 6, 7Michael |
| M 9/29 | Geometric Problems | Boyd ch 8Chetan |
| W 10/1 | Unconstrained Minimization | Boyd ch 9Satyaki |
| M 10/6 | Fall break |
| W 10/8 | Equality Constrained Minimization | Boyd ch 10 |
| M 10/13 | Interior Point Methods | Boyd ch 11Chetan |
| W 10/15 | Interior Point Methods | Boyd ch 11Satyaki |
| M 10/20 | BFGS | Nocedal ch 8Chetan |
| W 10/22 | BFGS implementation | Nocedal ch 9, Daume
| M 10/27 | Simulated Annealing | KirkpatrickLam |
| W 10/29 | Numerical Methods 1 | Press 10.0 - 10.3Yi |
| M 11/3 | Numerical Methods 2 | Press 10.4 - 10.6Yi |
| W 11/5 | Integer Programming | Nemhauser ch II.4Yi |
| M 11/10 | Machine Translation: MERT | Och 2003, Cer | Tag |
| W 11/12 | Making large scale SVM learning practical | Joachims | Michael |
| M 11/17 | Structured Max-Margin | Taskar 2003 Taskar 2004 |
| W 11/19 | SMO for Structured Max-Margin | Taskar |
| M 11/24 | Transductive SVM | Collobert |
| W 11/26 | Thanksgiving break |
| M 12/1 | Inducing features of random fields | Della Pietra | Tag |
| W 12/3 | Exponentiated Gradient | Bartlett | Satyaki |
| M 12/8 | Dirichlet Process Intro | Ranganathan | Lam |
| W 12/10 | Variational Inference for Dirichlet Process Mixtures | Blei | Lam |
Grading
- Lectures and participation: 60%
- Final project: 40%
gildea @ cs rochester edu
November 24, 2008
|