Causal Discovery

This book presents an overview of causal discovery , an emergent field with important developments in the last few years, and multiple applications in several fields.

The book is divided into three parts. The first part provides the necessary background on causal graphical models and causal reasoning. The second describes the main algorithms and techniques for causal discovery: (a) causal discovery from observational data, (b) causal discovery from interventional data, (c) causal discovery from temporal data, and (d) causal reinforcement learning. The third part provides several examples of causal discovery in practice, including applications in biomedicine, social sciences, artificial intelligence and robotics.

Topics and features:

• Includes the necessary background material: a review of probability and graph theory, Bayesian networks, causal graphical models and causal reasoning
• Covers the main types of causal discovery: learning from observational data, learning from interventional data, and learning from temporal data
• Illustrates the application of causal discovery in practical problems
• Includes some of the latest developments in the field, such as continuous optimization, causal event networks, causal discovery under subsampling, subject specific causal models, and causal reinforcement learning

• Provides chapter exercises, including suggestions for research and programming projects
  This book can be used as a textbook for an advanced undergraduate or a graduate course on causal discovery for students of computer science, engineering, social sciences, etc. It can also be used as a complement to a course on causality, together with another text on causal reasoning. It could also serve as a reference book for professionals that want to apply causal models in different areas, or anyone who is interested in knowing the basis of these techniques.

  The intended audience are students and professionals in computer science, statistics and

  engineering who want to know the principles of causal discovery and / or applied them in different

  domains. It could also be of interest to students and professionals in other areas who want to apply

  causal discovery, for instance in medicine and economics.

  Provides an accessible introduction to the field, including the foundations of causality and causal graphical models Includes numerous examples, exercises, and additional references Illustrates the application of causal discovery in different domains

  Autorentext

L. Enrique Sucar is Senior Research Scientist at the National Institute for Astrophysics, Optics and Electronics, Puebla, Mexico. He has published more than 400 papers in refereed journals and conferences, and is author of the Springer book, Probabilistic Graphical Models (2021, 2nd ed.).

Inhalt

1. Introduction.- 2. Causality.- 3. Causal Graphical Models.- 4. Causal Discovery from Observational Data.- 5. Causal Discovery from Interventional Data.- 6. Causal Discovery in Time Series.- 7. Causal Reinforcement Learning.