11-711: Algorithms for NLP, Fall 2017
Taylor Berg-Kirkpatrick and Robert Frederking
Lecture: Tuesday and Thursday 1:30pm-2:50pm, DH 1212
Recitation: Friday 1:30pm-2:20pm, MM 103
Office Hours: Taylor - Friday 12pm-1pm, GHC 6403
Bob - by appointment
Hieu Pham, Akshay Srivatsan and Maria Ryskina
Office Hours: Hieu - Monday 2pm-3pm, GHC 6418
Akshay - Wednesday 2pm-3pm, GHC 6418
Maria - Thursday 3:30pm-4:30pm, GHC 6603
11/17/17: Slides from the tenth recitation posted here: EM Recitation Slides.
11/13/17: Project 4 has been released. It is due Dec 4 at 11:59pm ET.
11/9/17: There is no recitation or OH this Friday (Nov 10) because of 50th Anniversary celebration.
11/6/17: Project 3 now due Sunday 11/12 by 11:59pm.
11/3/17: Slides from the ninth recitation posted here: P3 Interface Recitation Slides.
10/30/17: Project 3 has been released. It is due Nov 10 at 11:59pm ET.
10/28/17: Notes from the eighth recitation (coarse-to-fine part) posted here: Coarse-to-fine Recitation Notes.
10/23/17: Parsing scores for a correctly implemented parser have been added to the Project 2 page.
10/22/17: Project 2 now due Monday 10/30 by 11:59pm.
10/16/17: There is no recitation or OH this Friday (Oct 20) because of mid-semester break.
10/14/17: Taylor will hold extra office hours this week on Wed at 12:30pm at GHC 6403 (Taylor's office).
10/10/17: Project 2 has been released. It is due Oct 27 at 11:59pm ET.
10/7/17: Slides from the sixth recitation posted here: PCFG Recitation Slides. (Powerpoint version included for fans of the animation!)
9/29/17: Slides from the fifth recitation posted here: CRF Recitation Slides. (typos fixed 10/2)
9/28/17: Notes from the fourth recitation posted here: HMM Recitation Notes.
9/22/17: Slides from the fourth recitation posted here: HMM Recitation Slides.
9/21/17: A sample writeup (but for a different assignment) is available here: Sample writeup.
9/21/17: Canvas is up. See project submission instructions below.
9/19/17: Project 1 now due Saturday 9/23 by 11:59pm. Submission details to follow
9/15/17: Slides from the third recitation posted here: Implementation Tricks Slides.
9/8/17: Notes from the second recitation posted here: KN Recitation Notes. (updated 9/10)
9/1/17: Project 1 has been released.
9/1/17: Slides from the first recitation posted here: Project Setup Recitation Slides.
8/31/17: Piazza link posted above.
8/25/17: First lecture will be on Tuesday 8/29 at 1:30pm in DH 1212.
This course will explore current statistical techniques for the automatic
analysis of natural (human) language data. The dominant modeling paradigm is
corpus-driven statistical learning, with a split focus between supervised and
unsupervised methods. This term we are making Algorithms for NLP a lab-based course. Instead
of homeworks and exams, you will complete four hands-on coding projects.
This course assumes a good background in basic probability and a strong ability
to program in Java. Prior experience with linguistics or natural languages is
helpful, but not required. There will be a lot of statistics, algorithms,
and coding in this class.
Slides, materials, and projects for this new iteration of Algorithms for NLP are mainly borrowed from Dan Klein at UC Berkeley.
Submit projects using the class Canvas site.
1. Prepare a directory named 'project' containing no more than 3 files: (a) a jar named 'submit.jar', (b) a pdf named 'writeup.pdf', and (c) an optional jar named 'best.jar'. The jar named 'submit.jar' should contain your implementation of the core project that passes the basic requirements. For example, for project 1, the jar named 'assign1-submit.jar' is all that you would need to turn in -- renaming it 'submit.jar'. The pdf 'writeup.pdf' should contain your writeup for the project. Finally, the file 'best.jar' is an optional additional jar that implements the core project, but need not pass spot-checks. Include this last jar if you wish to demonstrate an improvement over the basic project, possibly using approximations are alternative models.
2. Compress the 'project' directory you created in the last step using the command 'tar cvfz project.tgz project'.
3. Click on the assignments tab of the main Canvas course site and select the assignment corresponding to the project (e.g. Assignment 1 corresponds to Project 1). Click 'Submit assignment' button to open submission portal, then click 'Choose file' and select your compressed project directory 'project.tgz' created in the previous step. Finally, click the 'Submit assignment' button below.
Projects out of 10 points total:
6 Points: Successfully implemented what we asked
2 Points: Submitted a reasonable write-up
1 Point: Write-up is written clearly
1 Point: Substantially exceeded minimum metrics
Extra Credit: Did non-trivial extension to project
Late Day Policy
Each student will be granted 5 late days to use over the duration of the semester. There are no restrictions on how the late days can be used (e.g. all 5 could be used on one project.) Using late days will not affect your grade. However, projects submitted late after all late days have been used will receive no credit. Be careful!
The primary recommended texts for this course are:
Note that M&S is free online (may need to setup proxy). Also, make sure you get the purple 2nd edition of J+M, not the white 1st edition.
Note to StudentsTake care of yourself! As a student, you may experience a range of challenges that can interfere with learning, such as strained relationships, increased anxiety, substance use, feeling down, difficulty concentrating and/or lack of motivation. All of us benefit from support during times of struggle. There are many helpful resources available on campus and an important part of having a healthy life is learning how to ask for help. Asking for support sooner rather than later is almost always helpful. CMU services are available, and treatment does work. You can learn more about confidential mental health services available on campus at: http://www.cmu.edu/counseling/. Support is always available (24/7) from Counseling and Psychological Services: 412-268-2922.
Syllabus [subject to substantial change!]
|1||Aug 29||Course Introduction||J+M 1, M+S 1-3|
|Aug 31||Language Modeling I||J+M 4, M+S 6, Chen & Goodman, Interpreting KN||P1: Language Modeling (Due Sept 23)|
|2||Sept 5||Language Modeling II||Massive Data, Bloom, Perfect, Efficient LMs|
|Sept 7||Language Modeling III|
|3||Sept 12||Speech Recognition I||J+M 7|
|Sept 14||Speech Recognition II||J+M 9, Decoding|
|4||Sept 19||Speech Recognition III, HMMs|
|Sept 21||POS Tagging, NER, CRFs||J+M 5, Brants, Toutanova & Manning|
|5||Sept 26||Parsing I||M+S 3.2, 12.1, J+M 13|
|Sept 28||Parsing II||M+S 11, J+M 14, Best-First, A*, Unlexicalized|
|6||Oct 3||Parsing III||Split, Lexicalized, K-Best A*, Coarse-to-fine|
|Oct 5||Formal Grammar|
|7||Oct 10||Parsing IV||P2: PCFG Parser (Due Oct 30)|
|Oct 12||Parsing IV -- continued|
|8||Oct 17||Parsing V|
|Oct 19||Parsing VI|
|9||Oct 24||Structured Classification I||Pegasos, Cutting Plane|
|Oct 26||Structured Classification II||J+M 16, 18, 19, Adagrad, Subgradient SVM|
|10||Oct 31||Guest lecture (Bhiksha Raj)||P3: Discriminative Reranker (Due Nov 12)|
|Nov 2||Unsupervised transcription of language and music|
|11||Nov 7||Machine Translation: Alignment I||J+M 25, IBM Models, HMM, Agreement, Discriminative|
|Nov 9||Machine Translation: Alignment II|
|12||Nov 14||Machine Translation: Phrase-Based||Decoding|
|Nov 16||Machine Translation: Syntactic||Hiero, String-Tree, Tree-String, Tree-Tree|
|13||Nov 21||Morphology; Features and Unification||J+M 3, J+M 15 (Note: errors in textbook)|
|Nov 23||Thanksgiving Day|
|14||Nov 28||Semantics and Discourse I||J+M 17, J+M 18 (Note: errors in textbook)|
|Nov 30||Semantics and Discourse II||J+M 12.7.2|
|15||Dec 5||Semantics and Discourse III||J+M 19, 20.6-20.9|
|Dec 7||Semantics and Discourse IV||J+M 21|