The PhD program offered a series of diverse educational activities to support the academic development of doctoral students. Compared to previous years, the seminar offerings for the PhD program are particularly enriched and include important contributions from prominent international figures in the field of computer science.
The activities offered include "Advanced thematic seminars" in which international experts teach short courses on emerging topics of common interest. Overall, these programs support students in exploring different themes in computer science. Alongside the thematic seminars, there are numerous PhD Schools that doctoral students from all years are encouraged to participate in, which address more cross-cutting issues and provide important networking opportunities with other members of the community. Additionally, doctoral students have the opportunity to participate in major conferences in their field as part of their training. Finally, it should be noted that students consistently achieve important milestones in research, such as the acceptance of papers at national and international scientific conferences, many of which are ranked as A+/A++ level, and are both financially and instrumentally supported to participate in them.
PhD schools and conferences attended by students:INFOCOM 2023
AAAI 2023
ProbNum School 2023
NLDL Winter School 2023
WONS 2023
SIGGRAPH Asia 2022
NeurIPS 2022
STAG 2022
IOTSMS 2022
AIxIA 2022
Visits to international research institutesUniversity of Cambridge
Oxford University
Imperial College London
University College London
Technion
Yale University
Universitat Pompeu Fabra
UC Berkeley
CoursesFor students from all three years, some mini-courses focused on studying specific and hot topics in computer science are proposed:
Who: Prof
Cygdem Beyan (Univ. of Verona)
Title: "Foundations of multimodal learning"
When: March 21st, 2PM-6PM
Where: https://univr.zoom.us/j/99177373356
Description:
Multimodal Learning, a subfield of machine and deep learning, is a multi-disciplinary research area focusing on integrating and modeling multiple modalities, such as acoustics, linguistics, and vision. This course delves into the fundamental concepts of multimodal learning, including alignment, fusion, joint learning, temporal learning, and representation learning. We will examine recent state-of-the-art techniques and focus on effective computational algorithms tailored for various applications.
Who: Prof
Daniele De SensiTitle: "Scalability and Collective Communication in Large-Scale Deep Learning"
Description:
Training deep learning models at scale requires efficient communication strategies to handle massive datasets and model sizes. This course introduces the fundamental challenges of large-scale deep learning and explores how collective operations impact performance. We will cover key collective algorithms, analyze their costs using the alpha-beta model, and discuss optimization strategies for large-scale systems. By the end of the course, participants will have a solid understanding of how to model, analyze, and optimize communication in distributed deep learning.
Length (hours): 6-8 hours
When:
- May 8 - 14.00-16.00
- May 9 - 14.00-16.00
- May 14 - 14.00-16.00
- May 15 - 14.00-16.00
Where: Computer Science Department & online (room and remote link will be shared later)
Who: Prof
Simone Melzi and Prof Filippo Maggioli
Title: "
Functional maps: a functional approach to the alignment of embeddings"
Length (hours): 24
When:
- May 29: 4 hours (in person)
- May 30: 4 hours (in person)
- June 5: 4 hours (in person)
- June 6: 4 hours (remotely)
- June 27: Hakaton (in person)
Where: Computer Science Department (room numbers and remote link will be shared later)
Who: Prof
Giuseppe PerelliTitle: "Game-Theoretic Approach to Synthesis"
Length (hours): 8-10
When:
- June 16: 10-12 (in person)
- June 17: 10-12 (in person)
- June 18: 10-12 (in person)
- June 19: 10-12 (in person)
- June 20: 10-12 (in person)
Where: Computer Science Department (room and remote link will be shared later)
Who: Prof
Simone ScardapaneTitle: "Into the land of automatic differentiation"
Length (hours): 20
When: July 1st to July 7th 2025
Where: Computer Engineering Department
Who: Prof
Maurizio ManciniTitle: "Fundamentals of Video Game Programming in Unity 3D"
Length (hours): 20
When: Mid-July 2025
Where: Computer Science Department & online
Who:
Salvatore PontarelliTitle: "Hardware Programming"
Length (hours): 5
When: September 24th 2025
Where: Computer Science Department
Who: Prof
Danilo Avola and Prof
Daniele PannoneTitle: "Beyond Sight: Using Wi-Fi Sensing Techniques to Solve Computer Vision Tasks"
Length (hours): 4
When: TBA
Where: Computer Science Department
Didactic paths for high-level training in computer scienceThe PhD program in Computer Science has proposed numerous highly specialized and training-oriented seminars on numerous hot topics in collaboration with prominent national and international professors in the field.
The complete list of seminars offered is visible in the appropriate section.
Other educational activitiesEach doctoral student carries out educational activities both in the teaching and tutoring of undergraduate and graduate courses or workshops/seminars. These activities encourage students to acquire transversal knowledge, including communication skills.
Doctoral students are daily engaged in the development, reporting, and creation of documentation for external and internal projects in the department. In particular, some of them have had the opportunity to get to know working environments at the industrial or governmental level, increasing their skillness in managing systems that are not purely academic.
Among the educational activities, doctoral students often have the opportunity to follow and guide undergraduate and graduate thesis students, both in development and in the drafting of the document. This activity allows them to obtain the recognition of "Co-supervisor".