This schedule is tentative and subject to change as the semester progresses.
| Date | Lecture | Reading | Assignment | ||
|---|---|---|---|---|---|
| Introduction, N-grams | |||||
| Th | 1/11 | 1 | What is NLP? | SLP2 ch. 1 | A0: The Basics |
| Tu | 1/16 | 2 | Text Processing, Tasks, and Corpora; ChatGPT activity on Zoom due to 🌨❄⛄ | SLP3 2.0–2.3, NLTK Book 3.2–3.10 | |
| Th | 1/18 | 3 | N-gram language models | SLP3 3.0–3.5.2; Goldwater probability review through section 4 | A0 due |
| Tu | 1/23 | 4 | N-gram language models contd. | A1: N-Gram Language Models | |
| Th | 1/25 | 5 | Overview of Linguistics | Features English is missing - but most other languages have; SLP2 1.1, SLP3 13.1 | |
| Classification | |||||
| Tu | 1/30 | 6 | Classification: naïve Bayes; Noisy Channel Model | SLP3 4.0–4.5 | |
| Th | 2/1 | 7 | Lexical semantics: senses, relations, classes | SLP3 23.0–23.3 | |
| Fr | 2/2 | A1 due | |||
| Tu | 2/6 | 8 | Linear models for classification: features & weights | SLP3 5.0–5.2 | A2: Perceptron |
| Th | 2/8 | 9 | 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) | |
| Tu | 2/13 | 10 | Linear models contd. | ||
| Sequential Prediction | |||||
| Th | 2/15 | 11 | Parts of speech | SLP3 8.0–8.3 | |
| F | 2/16 | A2 due | |||
| Tu | 2/20 | No class: classes follow Monday schedule | |||
| Th | 2/22 | 12 | POS tagging: HMMs | SLP3 8.4.0–8.4.3 Eisenstein Notes, 7.1: Part-of-speech tagging | |
| F | 2/23 | MIDTERM 1 | (See study guide worksheet) | ||
| Tu | 2/27 | 13 | Algorithms for HMMs (mainly Viterbi); BONUS: Discriminative tagging with the structured perceptron | SLP3 8.4.4–8.4.6, Appendix A; BONUS: Eisenstein Notes, 7.5 (no need to read beyond structured perceptron); Neubig slides | |
| Th | 2/29 | 14 | Annotation; Universal POS annotation activity (tagset) | A3: HMM | |
| M-F | 3/4-3/8 | No class or office hours: Spring Break | |||
| Distributed Representations and Neural Networks | |||||
| Tu | 3/12 | 15 | Distributional representations and similarity | SLP3 ch. 6 | |
| Th | 3/14 | 16 | Deep learning and neural networks | SLP3 7.0–7.1, 7.3–7.5 | |
| F | 3/15 | A3 due | |||
| Tu | 3/19 | 17 | Neural sequence modeling with RNNs | SLP3 9.1–9.8; BONUS: Details of LSTMs | A4: LSTMs |
| Hierarchical Sentence Structure | |||||
| Th | 3/21 | 18 | Final project options. English syntax, CFGs | SLP3 17.0–17.2, skim 17.3 | |
| Tu | 3/26 | 19 | Syntax contd.; project: review task options and submit preferences | P0: Submit project team with topic | |
| Th-M | 3/28-4/1 | No class or office hours: Easter Break | |||
| Tu | 4/2 | 20 | (P)CFG parsing: Parsing as search; CNF. | ||
| Th | 4/4 | 21 | (P)CFG parsing contd.: CKY walkthrough | SLP3 17.5–17.8, Appendix C C.0-C.4 | |
| F | 4/5 | A4 due | |||
| Tu | 4/9 | 22 | (P)CFG parsing contd. | ||
| Th | 4/11 | 23 | In-class project work session and team meetings with course staff | P1: 1-2 page proposal due | |
| Tu | 4/16 | 24 | Dependency parsing | SLP3 18.0-18.2, 18.4 | A5: Syntax |
| Sequence-to-Sequence, Translation, and Other Applications | |||||
| Th | 4/18 | 25 | Neural sequence-to-sequence models | SLP3 9.7–9.8, 10.0–10.2 | |
| Tu | 4/23 | 26 | Guest Lecture: Computational pragmatics (Brandon Waldon) | P2: Progress update, including lit review, due | |
| Th | 4/25 | 27 | Text generation (mainly QA, summarization, MT; mention dialog, image captioning) (Shabnam Behzad) + Wrap-up | SLP3 13, 14.2, 14.4–14.7 | |
| F | 4/26 | A5 due | |||
| Tu | 4/30 | 28 | Project presentations | ||
| F | 5/3 | MIDTERM 2 | Study guide to appear | ||
| W | 5/8 | No course meeting during final exam slot | PROJECT REPORT DUE | ||
| Visit the GUCL website for NLP talks this semester and beyond! | |||||