Neural Networks and Deep Learning - Trustworthy AI and Applications - III Module (20 hours)

📖 Course Overview

The course is a module of NNDL course. It focuses on topics and challenges related to trustworthiness in specific cyber-physical contexts, such as driving and autonomous systems. It also explores how to make neural networks more interpretable, robust, and secure.

The main topics of the course are organized as follows:

  1. Explainable and Interpretable AI
  2. Anomaly and Out-of-Distribution Detection Methods
  3. Domain Generalization and Domain Adaptation
  4. Adversarial Robustness and Defense Methods
  5. Simulation and Tools for Trustworthy AI
  6. Functional Components in Autonomous Driving

🛠️ Format & Exam

  • Lectures: 20 hours (remote).
  • Exam (2 CFU): Project/Research work + oral discussion.

Based on the instructions provided in the NNDL course. Please register here by December 15th to access the lecture channel and receive notifications about any changes to the schedule. If you need a certificate of attendance, please be aware that the number of certified hours will be based strictly on the time recorded by Teams.

Lectures will be opened to everybody and will be given online on the following channel.

🗓️ Schedule

  • Lecture 1 – Explainable and Interpretable AI – 27/04/2027 9:00–11:00 - Giulio Rossolini

  • Lecture 2 – Anomaly and out-of-distribution detection methods – 28/04/2027 9:00–11:00 - Giulio Rossolini

  • Lecture 3 – Domain generalization and domain adaptation – 04/05/2027 9:00 - 11:00 - Giulio Rossolini

  • Lecture 4 – Attention mechanisms in computer vision – 05/05/2027 9:00 - 11:00 - Giulio Rossolini

  • Lecture 5 – Adversarial attacks and defenses – 11/05/2027 9:00 - 11:00 - Federico Nesti

  • Lecture 6 – Real-world attacks and defenses – 12/05/2027 9:00 - 11:00 - Federico Nesti

  • Lecture 7 – Simulators for autonomous driving – 18/05/2027 9:00 - 11:00 - Federico Nesti

  • Lecture 8 – Architectural and simulation tools for trustworthy AI – 19/05/2027 9:00 - 11:00 - Federico Nesti

  • Lecture 9 – Functional components in autonomous driving – 25/05/2027 9:00 - 11:00 - Daniel Casini

  • Lecture 10 – The Autoware framework for autonomous driving – 26/05/2027 9:00 - 11:00 - Daniel Casini

📂 Lectures

You can find all lectures and notebooks in the following OneDrive folder: LINK

The folder is password-protected. Attendees can contact me directly to request access.

📬 Contact

giulio.rossolini@santannapisa.it

federico.nesti@santannapisa.it

daniel.casini@santannapisa.it