Graph Neural Network
- 3 ECTS credits
- Academic year 2026-2027
- DP00BE94-3001
Education information
Implementation date
24.08.2026 - 28.08.2026
Enrollment period
-
Education type
Field-specific studies
Alternativity of education
Optional
Location
Venue location
Please visit Peppi for updated information on venues at Linnamaa campus!
Enrollment and further information
Education description
1. Introduction to Graph Neural Networks
• Graph theory basics: nodes, edges, adjacency matrix
• Graph representations and types
• Introduction to message passing and graph convolutions
2. Core GNN Architectures
• Graph Convolutional Networks (GCN)
• Graph Attention Networks (GAT)
• GraphSAGE and Aggregation Mechanisms
3. Applications of GNNs
• Node classification
• Link prediction
• Graph classification
4. Advanced GNN Topics
• Heterogeneous and dynamic graphs
• Graph pooling and hierarchical GNNs
• Explainability and interpretability in GNNs
• Neuro-Symbolic reasoning in GNNs
5. Implementing GNNs
• Using PyTorch Geometric and DGL
• Model training, loss functions, and evaluation
• Best practices and debugging tips