Computer Science 248/448
Statistical Speech and Language Processing
Fall 2019
|
|
Instructor:
Dan Gildea (Office hours: Mon 2:00-3:00pm, Thu 2:00-3:00pm)
TA: Wasif Chowdhury (Office hours: Tue 2:00-3:00pm, Friday: 2:10-3:10pm)
Location: Tu/Th 11:05am-12:20pm, Gavet 312.
Homeworks
Recommended resources:
Syllabus
| On | we will cover | which means that after class you will understand | if before class you have read |
| 8/29 |
Perceptron for Tagging |
NLP = ML + DP
| Collins 2002 |
| 9/3 |
Hidden Markov Models |
viterbi, forward
| JM 6 |
| 9/5 |
CRFs |
f - E[f]
| McCallum 2001 |
| 9/10 |
Max Margin Learning |
subgradients
| Tsochantaridis 2004, Ratliff 2006 |
| 9/12 |
EM for HMM |
| JM 6 |
| 9/17 |
latent CRF |
| |
| 9/19 |
loss-aware CRF |
| Kakade 2002 |
| 9/24 |
Context Free Grammars |
highest prob string, infinite trees
| JM 13 |
| 9/26 |
Context Free Grammars |
viterbi, inside-outside
| |
| 10/1 |
Context Free Grammars |
viterbi, inside-outside
| |
| 10/3 |
Parsing |
minimum risk decoding, max-margin
| Finkel 08 |
| 10/8 |
Parsing |
crf parsing, neural parsing
| Durrett 2015 |
| 10/10 |
Weighted Deduction |
generalized outside computation
| Goodman 1999 |
| 10/15 |
Fall Break |
| |
| 10/17 |
Penn Treebank |
right node raising
| |
| 10/22 |
Search |
dijkstra, knuth, a*
| |
| 10/24 |
Review |
| |
| 10/29 |
Midterm |
|
|
| 10/31 |
Midterm Solutions |
| |
| 11/5 |
Beam search
| buckets
| |
| 11/7 |
Dependency Parsing
| transition systems
| Eisner and Satta 1999 |
| 11/12 |
Machine Translation |
Model 1
| Brown 1993 |
| 11/14 |
Neural Machine Translation |
attention
| Sutskever 2014, Bahdanau 2014 |
| 11/19 |
Neural Machine Translation |
BPE, BLEU
| Vaswani 2017, Papineni 2002, Sennrich 2016 |
| 11/21 |
Semantics |
word2vec
| Mikolov 2013, Pennington 2014 |
| 11/26 | Encoder Pretraining |
bert and elmo
| Elmo, Bert |
| 12/3 |
Semantics |
lambda calculus
| JM 20, Zettlemoyer 2005, Liang 2011 |
| 12/5 |
Something fun |
| |
| 12/10 |
Something fun |
| |
Final Exam: Thursday, December 19, 4-7pm, Gavet 312.
Grading
- Final exam: 35%
- Homeworks: 40%
- Midterm: 25%
- No late homework
- Programming assignments must be in python, and must run on class account
gildea @ cs rochester edu
November 26, 2019
|