General Information
Course Description
The aim of this course is to develop an understanding of the state-of-the-art techniques of neural networks and to apply them in practice, to natural language processing problems in particular. Thursday sessions will be typically dedicated to theory, Tuesday sessions – programming. During the practical sessions, we will use the PyTorch framework to implement our networks.
Script
The theoretical content can be found in the script (caution, frequent updates!).
Etherpad
for sharing code snippets etc.
Requirements
- BN: Complete the theoretical and the programming homework exercises. The homeworks will be published on this web page as we go.
- AP: Term paper based on a practical project: 4-5 pages for undergrad students, 7-10 pages for master students. Guidelines
Schedule
Time | Week | Content | Homework | Solutions |
---|---|---|---|---|
04.04.2023 | 01 | Introduction and overview | Software installation | — | — |
11.04.2023 | 02 | Tensors | Vektoren Matrizen | Lecture Code | Coding Ex_01 | Solution_01 |
18.04.2023 | 03 | Encoding and embedding | Intro Python | Coding Ex_02 | Solution_02 |
25.04.2023 | 04 | Building neural modules | Theory EX_01 | Solution |
02.05.2023 | 05 | Linear regression | LR_Sklearn | — | — |
09.05.2023 | 06 | Intro | Gradient Descent (Colab Zip) | Coding EX_03 (Colab Download) | Solution |
16.05.2023 | 07 | POS Tagging Continued (1 2) | Theory Ex_02 | — |
23.05.2023 | 08 | Dev Sets Data Analysis Batches Tensors (1 2) | — | — |
30.05.2023 | 09 | Tensors continued | Coding Ex_04 | Solution |
06.06.2023 | 10 | Contextualization | Coding Ex_05 | Solution |
13.06.2023 | 11 | LM from Scratch | Coding Ex_06 | Solution |
20.06.2023 | 12 | LM to POS tagger | Theory Ex_03 bis 29.6. | — |
27.06.2023 | 13 | LM to POS tagger 2 | project prep | — |
04.07.2023 | 14 | Tue: Huggingface + Transformers (Stanford Colab Tutorial) – Thu: final project discussion | tba | — |
11.07.2023 | 15 | MLM Scoring | — | — |