This schedule is tentative and subject to change as the semester progresses.
| Date | Lecture | Reading | Assignment | ||
|---|---|---|---|---|---|
| Introduction, N-grams | |||||
| Th | 1/9 | 1 | What is NLP? | SLP2 ch. 1 | A0: The Basics |
| Tu | 1/14 | 2 | Text Processing, Tasks, and Corpora; ChatGPT activity | SLP3 2.1–2.3, 2.6-2.7, NLTK Book 3.2–3.10 | |
| Th | 1/16 | 3 | N-gram language models | SLP3 3.1-3.4; Goldwater probability review through section 4 | A0 due |
| Tu | 1/21 | 4 | N-gram language models contd. | A1: N-Gram Language Models | |
| Th | 1/23 | 5 | Overview of Linguistics | Features English is missing - but most other languages have; SLP2 1.1, SLP3 13.1 | |
| Classification | |||||
| Tu | 1/28 | 6 | Classification: naïve Bayes; Noisy Channel Model | SLP3 4.1–4.6 | |
| Th | 1/30 | 7 | Lexical semantics: senses, relations, classes | SLP3 G.1–G.4 | |
| F | 1/31 | A1 due | |||
| Tu | 2/4 | 8 | Linear models for classification: features & weights | SLP3 5.1–5.3 | A2: Perceptron |
| Th | 2/6 | 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 | 2/11 | 10 | Parts of speech | SLP3 17.1–17.2 | |
| Sequential Prediction | |||||
| Th | 2/13 | 11 | POS tagging: HMMs | SLP3 17.4.1–17.4.4 Eisenstein Notes, 7.1: Part-of-speech tagging | |
| F | 2/14 | A2 due | |||
| Tu | 2/18 | No class: classes follow Monday schedule | |||
| Th | 2/20 | 12 | MIDTERM 1 | (See study guide worksheet in Canvas) | |
| Tu | 2/25 | 13 | Algorithms for HMMs (mainly Viterbi); BONUS: Discriminative tagging with the structured perceptron | SLP3 17.4.5–17.4.6, Appendix A; BONUS: Eisenstein Notes, 7.5 (no need to read beyond structured perceptron); Neubig slides | A3: HMM |
| Th | 2/27 | 14 | Annotation; Universal POS annotation activity (tagset) | ||
| M-F | 3/3-3/7 | No class or office hours: Spring Break | |||
| Distributed Representations and Neural Networks | |||||
| Tu | 3/11 | 15 | Distributional representations and similarity | SLP3 ch. 6 | |
| Th | 3/13 | 16 | Deep learning and neural networks | SLP3 7.1–7.3, 7.5–7.6 | A3 due |
| F | 3/14 | ||||
| Tu | 3/18 | 17 | XOR in neural networks. Neural sequence modeling with RNNs | SLP3 8.1–8.4; BONUS: Details of LSTMs | A4: LSTMs |
| Th | 3/20 | 18 | Final project options. Neural sequence-to-sequence models | SLP3 8.7, 9.1–9.2 | |
| Tu | 3/25 | 19 | Text generation (mainly QA, summarization, MT; mention dialog, image captioning); project: review task options and submit preferences | SLP3 13, 14.2, 14.4–14.7 | |
| Hierarchical Sentence Structure | |||||
| Th | 3/27 | 20 | Choose project topic as a team. Generation contd. | P0: Submit project team with topic | |
| M | 3/31 | A4 due | |||
| Tu | 4/1 | 21 | English syntax, CFGs | SLP3 18.1–18.3, skim 18.4 | P0: Submit project team with topic |
| Th | 4/3 | 22 | (P)CFG parsing: Parsing as search; CNF. | A5: Syntax | |
| Tu | 4/8 | 23 | In-class project work session and team meetings with course staff | P1: 1-2 page proposal due | |
| Th | 4/10 | 24 | (P)CFG parsing contd.: CKY walkthrough | SLP3 18.6-18.8, Appendix C C.1-C.4 | |
| Tu | 4/15 | 25 | Dependency parsing | SLP3 19.2, 19.4 | A5 due |
| Th-M | 4/17-4/21 | No class or office hours: Easter Break | |||
| Other Topics | |||||
| Tu | 4/22 | 26 | Guest Lecture: Computational pragmatics (Brandon Waldon) + Wrap-up | P2: Progress update, including lit review, due | |
| Th | 4/24 | 27 | MIDTERM 2 | Study guide to appear | |
| Tu | 4/29 | 28 | Project presentations | ||
| W | 5/7 | No course meeting during final exam slot | PROJECT REPORT DUE | ||
| Visit the GUCL website for NLP talks this semester and beyond! | |||||