DLAI-s2-2025
Deep Learning & Applied AI @Sapienza
Course material, 2nd semester a.y. 2024/2025, Dept. of Computer Science
News 🗞️
- 25/05/2025: The exam dates are now published. Scroll down for details.
- 13/05/2025: The list of projects has been published. Scroll down for details.
- 06/05/2025: Today’s seminar is also livestreamed here. Video recording will be available later!
- 12/04/2025: No lecture on April 21-22 due to Easter holidays.
- 06/04/2025: The theoretical lecture of 07 April is cancelled by Sapienza. Please watch the recorded video lecture (scroll down for link) as a replacement.
- 30/03/2025: This week, the theoretical lecture will take place on Tue (01 April). There will be lab both on Mon 31st and Tue 01st.
- 07/02/2025: The course website is online. Welcome to DLAI 2024/25! The course will start on Mon 3rd March.
Logistics 🧭
Lecturer: Prof. Emanuele Rodolà
Assistants: Dr. Daniele Solombrino, Dr. Giorgio Strano
When: Mondays 14:00–16:00 and Tuesdays 13:00–16:00
Where:
Physical classroom: Aula L1 - Castro Laurenziano (RM018-E01PTEL026)
There is no virtual classroom, and the lectures will not be recorded.
Pre-requisites 🔑
Python fundamentals; calculus; linear algebra.
Textbook, reading materials and video recordings 📖
Due to the continuously evolving nature of the topic, there is no fixed textbook as a reference. Specific material in the form of scientific articles and book chapters will be given throughout the lectures.
In addition:
- here you can find video recordings of 2025 lectures
- In the GitHub pages of the previous academic years you can find longer versions of some video lectures
- here you can find video recordings about topics that may come in handy when studying deep learning
- here you can find some supplementary course notes
- here you can find annotated slides
- here you can find Anki flashcards on theoretical and practical lectures
Thanks to past and present students for the kind contributions!
Accessibility 👁️🗨️: Since last year, in an effort to create a more inclusive and accessible learning environment, all slides have been re-designed with readability in mind to support students with specific learning disabilities. We aim to ensure that everyone, regardless of learning differences, has equal access to the educational content provided. Should you need additional accommodations or have suggestions for further improving accessibility, please feel free to reach out.
Grading 📊
Exam dates
Each exam session has cutoff dates for submitting the project.
Only projects that are (i) submitted before the cutoff date and (ii) registered on Infostud will be graded.
- ~18 June 2025 - cutoff date 11 June~
- 15 July 2025 - cutoff date 08 July
- 03 September 2025 - cutoff date 27 August
Evaluation proceeds according to the following steps:
- A project (mandatory, accounts for 100% of the final grade)
- An oral exam (optional, attributes at most 3 points, added to or subtracted from the final grade)
We may require an oral exam in doubtful cases or whenever necessary.
- Project list: Check the google doc. Access to the document is restricted to students with a @studenti.uniroma1.it account; you must request access to it when you click the link.
- The template for the final project report is here.
- Please read the material above carefully!
Past exams 📑
Exam is now project-only, but you can still find some past theoretical questions here
Lectures 🗣️
| Date | Topic | Reading | Code & Data |
|---|---|---|---|
| Mon 03 Mar | Introduction | slides | |
| Tue 04 Mar | Data, features, and embeddings | slides ; linear algebra recap ; matrix notes | |
| Mon 10 Mar | Linear regression, convexity, and gradients | slides | |
| Tue 11 Mar | Tensor basics and Tensor operations |  |
|
| Mon 17 Mar | Overfitting and going nonlinear | slides | |
| Tue 18 Mar | Linear models and Pytorch Datasets | |
|
| Mon 24 Mar | Stochastic gradient descent | slides | |
| Tue 25 Mar | Logistic Regression and Optimization | |
|
| Mon 31 Mar | Multi-layer perceptron and back-propagation | slides | |
| Tue 01 Apr | Autograd and Modules | |
|
| Mon 07 Apr | Convolutional neural networks | slides ; video | |
| Tue 08 Apr | Convolutional neural networks | |
|
| Mon 14 Apr | Regularization, batchnorm and dropout | slides | |
| Tue 15 Apr | Uncertainty, regularization and the deep learning toolset ; Batchnorm and dropout | ; |
|
| Mon 21 Apr | Easter holidays | ||
| Tue 22 Apr | Easter holidays | ||
| Mon 28 Apr | PCA and VAEs | slides | |
| Tue 29 Apr | Variational Autoencoders | |
|
| Mon 05 May | VQ-VAE | |
|
| Tue 06 May | Tools of the trade | slides | |
| Mon 12 May | Adversarial learning & Geometric deep learning | slides | |
| Tue 13 May | Adversarial attacks, CycleGAN (optional), and Geometric deep learning (optional) | ; |
|
| Mon 19 May | Diffusion models | notes ; source | |
| Tue 20 May | Diffusion models | notes | |
| Mon 26 May | Self-attention and transformers | slides | |
| Tue 27 May | Seminar, Donato Crisostomi: Model Merging - What, Why, and How |
End