Lectures for INFO8010 Deep Learning, ULiège, Spring 2025.
- Instructor: Gilles Louppe
- Teaching assistants: François Rozet, Yann Claes, Victor Dachet
- When: Spring 2025, Friday 8:30 AM
- Classroom: B28 / Mania Pavella amphitheater
- Discord: https://discord.gg/5yZqTZhXFW
| Date | Topic |
|---|---|
| February 14 | Course syllabus [PDF] [video] Lecture 0: Introduction [PDF] [video] Lecture 1: Fundamentals of machine learning [PDF] [video] |
| February 21 | Lecture 2: Multi-layer perceptron [PDF] [video] [code 1, code 2] |
| February 28 | Lecture 3: Automatic differentiation [PDF] [video] [code] |
| March 7 | Lecture 4: Training neural networks [PDF] [video] |
| March 14 | Lecture 5: Convolutional neural networks [PDF] [video] [code] |
| March 21 | Lecture 6: Computer vision [PDF] [video] |
| March 28 | Lecture 7: Attention and transformers [PDF] [video] |
| April 4 | Code: GPT, from scratch! Lecture 8: LLMs and foundation models [PDF] |
| April 11 | Lecture 9: Graph neural networks [PDF] |
| April 18 | Lecture 10: Uncertainty [PDF] [video] |
| May 9 | Lecture 11: Auto-encoders and variational auto-encoders [PDF] [video] [code] |
| May 16 | Lecture 12: Diffusion models [PDF] |
The goal of these two assignments is to get you familiar with the PyTorch library. You can find the installation instructions in the Homeworks folder. Each homework should be done in groups of 2 or 3 (the same as for the project) and must be submitted before 23:59 on the due date. Homeworks should be submitted on Gradescope.
- Homework 1: Tensor operations,
autogradandnn. Due by TBD. - Homework 2: Dataset, Dataloader, running on GPU, training a convolutional neural network. Due by TBD.
Homeworks are optional. If submitted, each homework will account for 5% of the final grade.
See instructions in project.md.
Due to progress in the field, some of the lectures have become less relevant. However, they are still available for those who are interested.
| Topic |
|---|
| Recurrent neural networks [PDF] [video] |
| Generative adversarial networks [PDF] [video] |