Dealing with Large Networks: Temporal Graph Learning and Clustering

Day - Time: 14 October 2026, h.11:00

Place: Area della Ricerca CNR di Pisa - Room: C-29

Speakers

• Nelson Aloysio Reis de Almeida Passos

Referent

Giulio Del Corso

Abstract

How to aggregate information from nodes, edges, and temporal dynamics is a common task known as graph clustering or pooling in machine learning on graphs, or community detection in network science. Graph neural networks have displayed state-of-the-art performance on many downstream tasks, but their advantage over established algorithms is far less settled, especially under demands of efficiency and recovery accuracy. We talk a bit about some of the principles, primitives, and pooling techniques for getting around this task.