Introduction to Network Inference and Network Learning in R: Markov random Fields and Bayesian Networks

Unlock the Power of Network Inference Methods in the Real World, All in One Exciting Day! In this introductory course, I will provide a brief overview of the two types of graphical models: undirected (Markov random fields) and directed (Bayesian networks). I will explain how these models can be applied in the real world and how to estimate (and interpret) the dependencies between variables of interest (in the form of a network) using various inference methods. We will also explore practical examples (using real data) and exercises (using R) to reinforce your learning.

Researchers, students, engineers, and analysts.

For an overview of all our Winter school courses offered by the Department of Methodology and Statistics please click here.

The aim of this course is to provide fundamental knowledge of graphical models for network inference for those people such as medical scientists, psychologists, biologists, economists, and other researchers who wish to learn the network structure for discovering the relationships between variables of interest.

By the end of this course, students will be able to:

• Understand Graphical Models: Gain a sufficient understanding of Graphical Models (Bayesian Networks and Markov Random Fields) and their real-world applications across various fields.
• Interpret Network Structures: Learn how to interpret network structures, including the meaning of nodes, edges, (in)dependency, and (blocked) paths. Discover how these relate to the dependencies between variables and their impact on the presence or absence of edges.
• Network Structure Estimation: Explore various network inference methods and acquire the skills to estimate undirected and directed network structures. These structures illustrate the dependencies between variables of interest, and you will use the R programming language for this purpose.
• When to Use Which Method: Compare Markov Random Fields and Bayesian Networks and gain insights into when to choose each method.

This one-day introductory course is dedicated to only network inference, where nodes represent variables, facilitating the discovery of relationships between variables in the form of a network. It serves as a part of the comprehensive ‘Network Science’ course offered at Utrecht University’s summer school, covering a wide range of networks, network analysis and network modeling.