Graph neutral networks (GNN) are a prominent class of deep learning methods designed for tasks whose inputs are graphs. Many natural and social phenomena can be viewed as graphs that contain rich relation (link) information among elements (nodes). GNN focus on a unique non-Euclidean data structure for machine learning, performing tasks such as node classification, link prediction, and clustering.

The merit of GNN lies in its capacity to represent data in a graph-structured way that can reveal valuable information emerging from a higher-dimensional representation of nodes and their links. On the other hand, idopNetworks represent the most informative and precise graphs that encapsulate all nodes and their links as a cohesive whole. A natural question arises: what happens if idopNetworks are embedded into GNN?

Now, a group of dedicated researchers cannot wait to sit together to discuss this question. They forget holidays, weekends, and even dinners to explore this exciting and interesting unknown.

Time: 17:00-20:00 Dec. 27, 2024

Venue: A6-101

Zoom: 559 700 6085

Password: BIMSA

Open remark

Rongling Wu

Presentations

Yaqing Wang

Shuaikang Ma

Discussion

Yunfeng Cai

Yishuai Niu

Mingming Sun

Yaqing Wang

Jie Wu

Rongling Wu