Home
Graph and geometric machine learning are fields that combine graph representations for data and machine learning. The goal is to develop machine learning models that can learn from graphs and make predictions based on the relationships between nodes.
Graph Machine Learning (GML) is a broad field with many use case applications and comprising multiple different supervised and unsupervised ML tasks. One of the primary purposes of GML is compressing large sparse graph structures while maintaining important signals for prediction and inference.
Geometric machine learning, or geometric deep learning (GDL), is a field of deep learning that aims to generalize neural network models to non-Euclidean domains. These domains include graphs and manifolds.
This workshop provides graduate students with the necessary skills for understanding and applying graph machine learning techniques. Among the covered topics, you will find the fundamentals of graph theory, practical applications of graph neural networks, and advanced methods for graph-based data analysis.
- Repo: https://github.com/ua-datalab/GraphML
- YouTube Playlist
- Mondays at 2PM: Weaver Science and Engineering Library Rm 212.
- Zoom: https://arizona.zoom.us/j/86423223879
- Qualtrics Registration: Link
| Date | Topics Covered | Instructor | Helpers | Code / Notebook |
|---|---|---|---|---|
| 04/01/24 | Graph ML Part-1 Why Graph ML and basics of graph theory |
Shashank | Carlos |
Colab Notebook  YouTube Recording |
| 04/08/24 | Graph ML Part-2 Node representations: Deepwalk and node2vec |
Shashank | Carlos |
Colab Notebook YouTube Recording |
| 04/15/24 | Graph ML Part-3 Basics of GNN - Node classification |
Shashank | Carlos |
Colab Notebook [YouTube Recording] |
| 04/22/24 | Graph ML Part-4 | Shashank | Carlos | Introduction to Graph Convolutions |
| 04/29/24 | Graph ML Part-5 | Shashank | carlos | Application of Graph Convolutions Networks |
