Neural Networks for Data Science Applications
Master's Degree in Data Science (2023-2024)
For the previous year (2022-2023), refer to this page.
Important Info
| Material: slides, assignments, and grading will be done via Google Classroom. | |
| Timetable: Wednesday 5-7 PM (Aule A5 + streaming in A6, Via Ariosto), Friday 8-11 (Aule A5 + streaming in A6, Via Ariosto). |
News
- On December 1st we will have a seminar on unifying neural representations.
- Classes will start on September 27th (see the faculty calendar).
General overview
The course provides a general overview on neural networks as compositions of differentiable blocks, that are optimized numerically. We describe common building blocks including convolutions, self-attention, batch normalization, etc., with a focus on image, audio, and graph domains. The course combines rigorous mathematical descriptions with many coding sessions in TensorFlow.
More information about the topics, the exam, organization, etc. can be found on the introductory slides.
Teaching assistants
Material
Lab sessions (mandatory) implemented in TensorFlow are in blue. Homeworks and projects (mandatory) are in red. Seminars (optional) are in green.
| Â | Date | Content | Material |
|---|---|---|---|
| L0 | 27/09 | About the course | Slides Video |
| L1 | 27/09 | Introduction | Slides Video |
| L2 | 29/09 | Preliminaries | Slides Video (1/2) Video (2/2) |
| L3 | 4/10, 6/10 | Linear models | Slides Video (1/3) Video (2/3) Video (3/3) |
| - | 6/10 | The AI software ecosystem (basics) | Slides |
| E1 | 11/10 | Lab session: logistic regression from scratch | Notebook |
| L4 | 13/10 | Fully-connected models | Slides Video (1/2) Video (2/2) |
| L5 | 18/10, 20/10 | Automatic differentiation | Slides Video (1/2) Video (2/2) |
| L6 | 25/10, 27/10 | Convolutional neural networks | Slides Video |
| L7 | 27/10 | Convolutions beyond images | Slides Video (1/2) Video (2/2) |
| E2 | 07/11 | Lab session: building CNNs with the Functional API | Notebook |
| H1 | - | Homework: Saliency maps for interpretability | Template Video |
| L8 | 10/11, 15/11 | Building deep convolutional networks | Slides Video (1/3) Video (2/3) Video (3/3) |
| E3 | 17/11 | Lab session: text classification with 1D CNNs | Notebook Video (1/3) Video (2/3) Video (3/3) |
| L9 | 29/11 | Attention-based neural networks (transformers) | Slides Video (1/3) Video (2/3) Video (3/3) |
| S1 | 01/12 | Unifying Representations in Neural Models (Donato Crisostomi, Marco Fumero) |
Slides (1/3) Slides (2/3) Slides (3/3) Notebook Video (1/2) Video (2/2) |
| L10 | 13/12 | Transfer learning | Slides Video |
| E4 | - | Lab session: Keras 3.0, JAX, and einops (optional) | Notebook |
| H2 | - | Homework 2: advanced transfer learning | Template Video |