PHD COLLOQUIA-2

Academic year
2024/2025 Syllabus of previous years
Official course title
PHD COLLOQUIA-2
Course code
PHD206-2 (AF:545149 AR:311533)
Modality
On campus classes
ECTS credits
2
Degree level
Corso di Dottorato (D.M.226/2021)
Educational sector code
INF/01
Period
Annual
Course year
1
Where
VENEZIA
Moodle
Go to Moodle page
The course will introduce the most recent Neural Network techniques to process non-euclidean data, such as graphs and meshes.
Students will gain a deep knowledge of the main challenges of developing and using Neural Network algorithms to graphs, together with state-of-the-art solutions from both theoretical and practical perspectives.
Basic knowledge of Linear Algebra, Probability, and Python Coding.
Challenges of graphs: permutation invariance and covariance
Convolution on graphs
Graphs Auto-Encoders
Hamilton, William L. Graph representation learning. Morgan & Claypool Publishers, 2020.
Materials provided during the lessons
Short paper, with the possibility to choose between:
- literature review paper
- application of GNNs to a specific problem
written
Lectures and hands-on sessions
English
Written exam
Definitive programme.
Last update of the programme: 27/01/2025