Difference between revisions of "Syllabus for Structured Prediction 10-710 in Fall 2011"

From Cohen Courses
Jump to navigationJump to search
Line 1: Line 1:
 
This is the syllabus for [[Structured Prediction 10-710 in Fall 2011]].  The slides available may be out of date (from the last version of the class) but will be updated after each lecture.
 
This is the syllabus for [[Structured Prediction 10-710 in Fall 2011]].  The slides available may be out of date (from the last version of the class) but will be updated after each lecture.
  
== August ==
+
== August/September ==
 +
 
  
* Tues 8/30. Intro (Noah)
 
* Thus 9/1. HMMs as structured-output prediction models. (William)
 
  
 
== September ==
 
== September ==
  
* Tues 9/6
+
* Tues 8/30. Intro (Noah)
* Thus 9/8
+
* Thus 9/1. HMMs as structured-output prediction models. (William)
* Tues 9/13
+
* Tues 9/6. HMMs to MEMMs. (Noah)
* Thus 9/15
+
* Thus 9/8. CRFs. (William).
* Tues 9/20
+
* Tues 9/13. Pseudo-likelihood inference. (William).
* Thus 9/22
+
* Thus 9/15. Conditional structure vs conditional estimation. (Noah).
* Tues 9/27
+
* Tues 9/20. Perceptrons and margin-based learning. (William).
* Thus 9/29 (Rosh Hashanah)
+
* Thus 9/22. Ranking perceptrons and margin-based learning. (William).
 +
* Tues 9/27. Generalizing SVM to structures. (Noah)
 +
* Thus 9/29. Dynamic programming techniques. (Noah; note this is Rosh Hashanah)
  
 
== October ==
 
== October ==
  
* Tues 10/4
+
* Tues 10/4. Stacking, Searn, and related "meta-learning" techniques. (William).
* Thus 10/6
+
* Thus 10/6. Test-time optimization with ILP and other methods (Noah).
* Tues 10/11
+
* Tues 10/11. PCFGs and parsing (Noah).
* Thus 10/13
+
* Thus 10/13. Spanning trees and dependency parsing. (Noah).
* Tues 10/18
+
* Tues 10/18. Kernels for structured inputs. (William).
* Thus 10/20
+
* Thus 10/20. Kernels for structured outputs. (Noah).
* Tues 10/25
+
* Tues 10/25. '''Mid-term project reports''' - short presentations of task, dataset, and baseline.
* Thus 10/27
+
* Thus 10/27. Learning edit distances. (William).
  
 
== November ==
 
== November ==
  
* Tues 11/1
+
* Tues 11/1. Weighted FSTs. (Noah).
* Thus 11/3
+
* Thus 11/3. Non-monotonic alignment for MT.(Noah).
* Tues 11/8
+
* Tues 11/8. EM for sequences. (William).
* Thus 11/10
+
* Thus 11/10. EM for trees. (Noah).
* Tues 11/15
+
* Tues 11/15. Discriminative latent-variable models. (Noah).
* Thus 11/17
+
* Thus 11/17. Regularizers derived from unannotated data. (William).
* Tues 11/22
+
* Tues 11/22. Posterior regularization and other EM variants. (Noah).
 
* Thus 11/24 -- ''Thanksgiving, no class''
 
* Thus 11/24 -- ''Thanksgiving, no class''
* Tues 11/29
+
* Tues 11/29. Classification on graphs (William).
  
 
== December ==
 
== December ==
  
* Thus 12/1
+
* Thus 12/1. Bayesian grammars (Noah).
* Tues 12/6
+
* Tues 12/6. Project presentations 1
* Thus 12/8
+
* Thus 12/8. Project presentations 2.
 +
* Mon 12/12. '''Final projects due.'''

Revision as of 11:13, 19 July 2011

This is the syllabus for Structured Prediction 10-710 in Fall 2011. The slides available may be out of date (from the last version of the class) but will be updated after each lecture.

August/September

September

  • Tues 8/30. Intro (Noah)
  • Thus 9/1. HMMs as structured-output prediction models. (William)
  • Tues 9/6. HMMs to MEMMs. (Noah)
  • Thus 9/8. CRFs. (William).
  • Tues 9/13. Pseudo-likelihood inference. (William).
  • Thus 9/15. Conditional structure vs conditional estimation. (Noah).
  • Tues 9/20. Perceptrons and margin-based learning. (William).
  • Thus 9/22. Ranking perceptrons and margin-based learning. (William).
  • Tues 9/27. Generalizing SVM to structures. (Noah)
  • Thus 9/29. Dynamic programming techniques. (Noah; note this is Rosh Hashanah)

October

  • Tues 10/4. Stacking, Searn, and related "meta-learning" techniques. (William).
  • Thus 10/6. Test-time optimization with ILP and other methods (Noah).
  • Tues 10/11. PCFGs and parsing (Noah).
  • Thus 10/13. Spanning trees and dependency parsing. (Noah).
  • Tues 10/18. Kernels for structured inputs. (William).
  • Thus 10/20. Kernels for structured outputs. (Noah).
  • Tues 10/25. Mid-term project reports - short presentations of task, dataset, and baseline.
  • Thus 10/27. Learning edit distances. (William).

November

  • Tues 11/1. Weighted FSTs. (Noah).
  • Thus 11/3. Non-monotonic alignment for MT.(Noah).
  • Tues 11/8. EM for sequences. (William).
  • Thus 11/10. EM for trees. (Noah).
  • Tues 11/15. Discriminative latent-variable models. (Noah).
  • Thus 11/17. Regularizers derived from unannotated data. (William).
  • Tues 11/22. Posterior regularization and other EM variants. (Noah).
  • Thus 11/24 -- Thanksgiving, no class
  • Tues 11/29. Classification on graphs (William).

December

  • Thus 12/1. Bayesian grammars (Noah).
  • Tues 12/6. Project presentations 1
  • Thus 12/8. Project presentations 2.
  • Mon 12/12. Final projects due.