This schedule is tentative and subject to change as the semester progresses.
| Date | Lecture | Reading | Assignment | ||
|---|---|---|---|---|---|
| Introduction, N-grams | |||||
| W | 1/8 | 1 | What is NLP? | SLP2 ch. 1 | A0: The Basics |
| M | 1/13 | 2 | Why do we need data?; Basic Text Processing (Regular Expressions, Normalization) | Working with text in Python 3, SLP3 2.0–2.3, NLTK Book 3.2–3.10 | |
| W | 1/15 | 3 | N-gram language models | SLP3 3.0–3.4.2; Goldwater probability review through section 4 | A0 due |
| M | 1/20 | No class or office hours: MLK Day | A1: Language Models | ||
| W | 1/22 | 4 | N-gram language models contd. | ||
| F | 1/24 | ||||
| M | 1/27 | 5 | Overview of Linguistics | Features English is missing - but most other languages have; SLP2 1.1, 25.1 | |
| Classification | |||||
| W | 1/29 | 6 | Classification: naïve Bayes; Noisy Channel Model | SLP3 4.0–4.5 | |
| F | 1/31 | A1 due | |||
| M | 2/3 | 7 | Lexical semantics: senses, relations, classes; Linear models for classification: features & weights | Appendix C.0–C.2, 5.0–5.1 | |
| W | 2/5 | 8 | Linear models for classification: discriminative learning (perceptron, SVMs, MaxEnt) | Daumé The Perceptron: 4.0–4.3; SLP3 4.7–4.8, ch. 5 (Further readings are suggested in slides) | A2: Perceptron |
| M | 2/10 | 9 | Linear models: discriminative learning (contd.); Parts of speech | SLP3 8.0–8.3; Eisenstein Notes, 7.1: Part-of-speech tagging | |
| Sequential Prediction | |||||
| W | 2/12 | 10 | POS tagging: HMMs | SLP3 8.4.0–8.4.3, 8.4.9 | |
| F | 2/14 | A2 due | |||
| M | 2/17 | No class or office hours: Presidents' Day (class on Tuesday instead) | |||
| Tu | 2/18 | 11 | MIDTERM EXAM | Study guide | |
| W | 2/19 | 12 | Algorithms for HMMs (mainly Viterbi) | SLP3 8.4.4–8.4.8, Appendix A | A3: HMM |
| M | 2/24 | 13 | Discriminative tagging with the structured perceptron; Annotation | Eisenstein Notes, 7.5 (no need to read beyond structured perceptron); Neubig slides | |
| W | 2/26 | 14 | Universal POS annotation activity (tagset) | ||
| Distributed Representations and Neural Networks | |||||
| A3 due | |||||
| M | 3/2 | 15 | Distributional representations and similarity | SLP3 ch. 6 | |
| W | 3/4 | 16 | Deep learning and neural networks | SLP3 7.0–7.1, 7.3–7.5 | |
| M-F | 3/9-13 | No class or office hours: Spring Break | |||
| M | 3/16 | 17 | Neural sequence modeling with RNNs (Michael) | SLP3 9.1–9.4 | A4: LSTMs |
| Hierarchical Sentence Structure | |||||
| W | 3/18 | 18 | English syntax, CFGs; forming project teams | SLP3 12.0–12.3, skim 12.4 | |
| F | 3/20 | P0: Suggest possible topics, form project teams | |||
| M | 3/23 | 19 | Syntax contd. | ||
| W | 3/25 | 20 | Syntax, CFGs contd. | ||
| F | 3/27 | A4 due | |||
| M | 3/30 | 21 | (P)CFG parsing | SLP3 13.0–13.2, 14.0–14.4 | P1: 1-2 page proposal due |
| W | 4/1 | 22 | (P)CFG parsing contd. | A5: Syntax | |
| Th-F | 4/2-3 | P2: groups meet with instructor & TA | |||
| M | 4/6 | 23 | Dependency parsing (includes bonus slides not covered in class) | SLP3 15.0–15.4.1, 15.6 | |
| W | 4/8 | 24 | Semantic role labeling | SLP3 16.0, ch. 20, pp. 1–10 | |
| F | 4/10 | P3: Progress update, including lit review, due | |||
| M | 4/9-13 | No class or office hours: Easter Break | |||
| Translation and Sequence-to-Sequence | |||||
| W | 4/15 | 25 | Statistical machine translation | SLP2 25.3–25.9 | |
| F | 4/17 | A5 due | |||
| M | 4/20 | 26 | Statistical machine translation contd. | ||
| W | 4/22 | 27 | Neural sequence-to-sequence models (Austin) | ||
| M | 4/27 | 28 | Last day of class; Context in language processing + Wrap-up | ||
| Th | 5/7 | VIRTUAL PROJECT POSTER SESSION (4:00-6:00 p.m.) | |||
| Sa | 5/9 | PROJECT REPORT DUE | |||
| Visit the GUCL website for NLP talks this semester and beyond! | |||||