Steam-powered Turing Machine University of Washington Department of Computer Science & Engineering
CSEP 517 - Natural Language Processing - Autumn 2013
Wed 6:30-9:20pm in MGH 231
  CSE Home  About Us    Search    Contact Info 

Instructor: Luke Zettlemoyer (lsz at cs dot washington dot edu)
Office hours: After class, until everyone leaves
TAs:
Yoav Artzi (yoav at cs dot washington dot edu)
Janara Christensen (janara at cs dot washington dot edu)
Anthony Fader (afader at cs dot washington dot edu)
Mark Yatskar (my89 at cs dot washington dot edu)
Office hours: Wed 5:30-6:30, CSE Third Floor Breakout (Open area by elevators)

Schedule [WARNING: topics subject to change!]

Week Dates Topics & Lecture Slides Notes (required reading) Textbook (recommended reading) Academic Papers (optional reading)
1 Sep 25 Introduction; Language Models (LMs) LM Notes J&M 4; M&S 6 [Smoothing] [Pitman-Yor] [Large LMs]
2 Oct 2 LM (cont.); Text Classification: Naive Bayes and Linear Models Naive Bayes (Sec. 1-4), Log-linear models (Sec 2) J&M 20-20.4; M&S 16.2-16.3 [Semi-supervised Naive Bayes] [MaxExt]
3Oct 9 Hidden Markov Models (HMMs) HMM Notes J&M 5.1-5.3, 6.1-6.4; M&S 9, 10.1-10.3
4 Oct 16 POS and Feature Rich Tagging CRF Notes [TnT Tagger] [Stanford Tagger] [SOTA POS]
5 Oct 23 POS (cont.); EM and Semi-supervised learning Naive Bayes (Sec. 4-6)
6 Oct 30 PCFGs and Parsing PCFG Notes Lexicalized PCFGs J&M 14; M&S 11-12 [Lexicalized] [Unlexicalized] [Split Merge]
7 Nov 6 Parsing(cont.); Word Alignment; Phrase-based MT IBM Models 1 and 2 Phrase-based Notes M&S 13; J&M 25.1-6;J&M 25.7-9 [IBM Models] [HMM Model] [Phrase-based Model] [Decoding] [Tuning]
8 Nov 13 Compositional Semantics J&M 17.1-4
9 Nov 20 Phrase-based MT (cont.); Syntax-based MT J&M 22.1-2 [Tree-to-String] [Hiero] [Tree-to-Tree] [Reordering]
10 Nov 27 No class. Happy Thanksgiving!
11 Dec 4 Discourse and Co-reference; TA Research Talks J&M 21.2-8

Textbooks

Contact

Homeworks

We will have 4 programming assignments, which will be listed below as they are assigned. All code and data for these assignments are available in the course ShareSpace. Look here for some hints and guidelines about how to do a good writeup. Please submit your assignments to the online DropBox.

Grading

The final grade will be based on the four homeworks (25% each). There will be no exams.

Course Administration and Policies


CSE logo Department of Computer Science & Engineering
University of Washington
Box 352350
Seattle, WA  98195-2350
(206) 543-1695 voice, (206) 543-2969 FAX