Natural Language Processing
Graduate course at the AI Graduate School, GIST. I served as the teaching assistant and ran the lab projects.
Instructor: Prof. Kangil Kim (TA. Hoyong Kim)
Term: Spring
Location: GIST, Gwangju
Time: Spring 2022
Lab projects
The lab track introduced practical skills complementary to the lectures, with six lab assignments over the semester.
- Lab 0. Environment setup (Google Colab, GIST GPU servers, Kaggle and GitHub submission)
- Lab 1. Representation of symbolic data
- Lab 2. NLP data preparation (midterm project)
- Lab 3. Problem formulation
- Lab 4. Encoder-decoder implementation
- Lab 5. Transformer implementation
All lecture materials are kept confidential to the course.
Schedule
| Week | Date | Topic | Materials |
|---|---|---|---|
| 1 | Mar 3 | Course introduction and programming basics Course overview and Python basics for the lab projects. | |
| 2 | Mar 8 - Apr 7 | Background of AI Overview of AI, problem solving, model representation, optimization and generalization, evaluation. | |
| 3 | Apr 7 - Apr 21 | Probabilistic and logic modeling Bayesian networks, Markov random fields, predicate calculus, inductive logic programming. | |
| 4 | Apr 26 - May 12 | Statistical relational learning Bayesian logic programming, Markov logic networks, probabilistic context-free grammars. | |
| 5 | May 17 - May 26 | Overview of natural language processing Linguistics essentials, NLP problems and their relations, n-gram models. | |
| 6 | May 31 - Jun 9 | Language models, machine translation, QA, and conversational systems Neural language models, encoder-decoder machine translation, question answering, dialogue systems. | |
| 7 | Jun 14 | Final exam Covers all lectures. |