Applications of Big Data and Machine Learning in Galaxy Formation and Evolution

Reviews / Votes

This book is an outstanding fusion of galactic astronomy and modern statistical analysis, including machine learning. The first half concisely covers fundamental processes such as radiation and gas dynamics, along with a wide range of galactic phenomena. The second half provides numerous practical examples, including both supervised methods like convolutional neural networks, as well as a strong emphasis on unsupervised techniques such as principal component analysis, VAE, and UMAP. Additionally, it explores statistical methods like copulas and advanced approaches such as topological data analysis, making it an indispensable resource for the big data era in astronomy.

Prof. Takeuchi, a pioneer in applying statistical methods to astronomy, has uniquely positioned this book at the intersection of galactic studies and modern data science. For graduate students eager to bridge these fields, this book eliminates the need for multiple textbooks, offering a singular, authoritative guide.

- Makoto Uemura, Hiroshima University, April 2025

Professor Takeuchi is a well-known researcher in extragalactic astrophysics and a recognized expert in statistical methods applied to this field. His expertise is so widely respected that even fellow astronomers frequently consult him on statistical questions related to astrophysics.

This book offers a comprehensive overview of galaxies and extragalactic astrophysics. As a teacher of astrophysics, I would strongly recommend it to students seeking a deeper understanding of galaxies and the physical equations that govern them across all redshifts. Beyond its solid theoretical foundation, the second part of the book delves into the application of statistics and data science in the study of galaxies. Readers will learn how to derive key physical and cosmological parameters necessary from data analysis and for gaining insights into galaxy evolution. It is an essential resource not only for students in extragalactic astrophysics but also for scientists interested in integrating data-science techniques into their research.

- Denis Burgarella, Laboratoire d'astrophysique de Marseille, May 2025