EE-452 / 4 credits

Teacher(s): Frossard Pascal, Thanou Ntorina

Language: English


Summary

Fundamentals, methods, algorithms and applications of network machine learning and graph neural networks

Content

Context

In the last decade, our information society has mutated into a data society, where the volume of worldwide data grows increasingly fast. An increasing amount of this data is structured on networks of different forms. How to make sense of such tremendous volume of data? Developing effective techniques to extract meaningful information from large-scale and high-dimensional network datasets has become essential for the success of business, government and science.

Objective

The goal of this course is to provide a broad introduction to effective methods algorithms in data science, network analysis and network machine learning. A major effort will be given to show that existing data analysis techniques can be defined and enhanced on graphs. Graphs can encode complex structures like cerebral connection, stock exchange, and social network. Strong mathematical tools have been developed based on statistics, or linear and non-linear graph spectral harmonic analysis to advance the standard data analysis algorithms. At the same time, modern machine learning tools such as neural networks have been adapted to process data defined on network structures. The objective of the class is to develop fundamentals and review algorithms that permit to develop modern network data analysis methods. The main topics of the course are networks, network data analysis, unsupervised and supervised learning on graphs and networks, graph generative models, sparse representation, multi-resolution analysis, graph neural networks.

Structure

The course is organized into two parts: lectures (2 hours) and lab assignements and projects (1 hour). The essential objective of the exercises and lab assignements is to apply the theory on real-world cases. The objective of the projects is to study practical network machine learning cases, and develop effective solutions based on tools studied in the class. 
Evaluation
Evaluation will be conducted on a continuous basis: homeworks and coding assignments.

 

Keywords

graph representation learning, machine learning, network science

Learning Prerequisites

Required courses

Fundamentals of Machine Learning, or equivalent

Signal Processing, or equivalent

Introduction to Statistics, or equivalent

Python programming

Learning Outcomes

By the end of the course, the student must be able to:

  • Apply modern machine learning techniques to network data
  • Analyze network properties, network data distributions, and properties of the most common network machine learning algorithms
  • Propose solutions for network data analysis problems

Transversal skills

  • Use a work methodology appropriate to the task.
  • Give feedback (critique) in an appropriate fashion.
  • Communicate effectively, being understood, including across different languages and cultures.

Resources

Moodle Link

In the programs

  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: During the semester (summer session)
  • Subject examined: Network machine learning
  • Lecture: 2 Hour(s) per week x 14 weeks
  • Project: 2 Hour(s) per week x 14 weeks
  • Type: optional

Reference week

Related courses

Results from graphsearch.epfl.ch.