Network machine learning
EE-452
Media
This file is part of the content downloaded from Network machine learning.
- Objectives: To provide an introduction to methods and algorithms in network analysis and machine learning. A major goal is to understand, analyze, and design network-based algorithms in the context of learning and representation of structured data.
- Prerequisites: linear algebra, statistics, calculus, digital signal processing or equivalent, machine learning, programming basics (python)
- Course Organisation: Combination of lectures and lab sessions (please come with your laptops!)
- Lab sessions on Mon 16-18 (AAC231) and lectures on Tue 13-15 (SG0211)
- Grading: midterm (40%) and project (60%)
- Course Teaching Team:
- Main instructors: Dr. Dorina Thanou and Prof. Pascal Frossard
- TAs: Manuel Madeira (lead TA), Abdellah Rahmani, Jeremy Baffou, Sevda Ogut, William Cappelletti, Yiming Qin
- Schedule:
- Important Note: Technical questions related to projects and notebooks should be asked during the exercise sessions or on the moodle Q&A.
- References:
- Network Science, Barabasi
- Graph Representation Learning, Hamilton
- Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges, Bronstein et al.
- Project Introduction (Folder)
- Notebook 3 (Folder)
- Lecture 4 - Learning embeddings on graphs: an unsupervised approach (File)
- GIF node2vec (File)
31 March - 6 April
- Notebook 4 (Folder)
- Midterm Preparation Questions (File)
- Lecture 6 - Graph neural networks: Building blocks (File)
7 April - 13 April
- Notebook 5 (Folder)
- Midterm Preparation Questions - Solutions (File)
- Lecture 7 - Graph neural networks: Main architectures (File)
28 April - 4 May
- Lecture 8 - Graph Transformers (w/ Prof. Xavier Bresson) (File)
- NML Midterm Solutions (File)
- NML Midterm Statistics (File)