The course is intended as a broad overview to neural networks, as used today in a number of applicative fields. It provides a strong theoretical and practical understanding of how neural networks and modern deep networks are designed and implemented, highlighting the most common components, ideas, and current limitations. We will review the general paradigm of building differentiable models that can be optimized end-to-end with gradient descent from data, and then overview essential components to design architectures able to work on images (convolutive layers), sequences (recurrent layers), and sets (transformer layers). The last part of the course will then focus on a selection of important research topics, including graph neural networks, continual learning, and generative models.
Calendar: 27 September - 21 December, 2022.
Class Schedule: Tuesday 13:00 - 15:00 Classroom 41 (San Pietro in Vincoli)
Thursday 11:00 - 15:00 Classroom 7 (San Pietro in Vincoli)
Attendance: In presence only, according to the latest university regulations.
Masks are still mandatory in class, and it is highly recommended to book a seat on Prodigit.
Visit the Sapienza website for further updates.
Material: Slides, notebooks, assignments, and news will be available on the course Classroom.
Teachers
Exams
Exams must be booked electronically via the INFOSTUD portal. Scheduled exam sessions for the year 2022/2023:
Session I: January 13, 2023 - Aula I, Via Vincenzo Gaglioti, Edificio CU032, 15:00
Session II: February 16, 2023 - Aula 10, Via Eudossiana 18, 15:00
Extraordinary Session I: March 29, 2023 - Aula A3, Via Ariosto, 17:00
Session III: June 23, 2023 - Aula A3, Via Ariosto, 10:00
Session IV: July 20, 2023 - Aula A3, Via Ariosto, 10:00
Session V: September 4, 2023 - Â Aula A3, Via Ariosto, 10:00
Extraordinary Session II: October 27, 2023 - Aula A6, 14:00