New Book Explores Agent-Based Modeling, Multi-Agent Systems

A new book provides an overview of agent-based modeling (ABM) and multi-agent systems (MAS), emphasizing their significance in understanding complex economic systems. The book's 14 chapters feature practical examples, applications, and case studies, with a focus on how learning, particularly machine learning, can be integrated into multi-agent systems to enhance the adaptation and behavior of agents in dynamic environments.

The book is titled Machine Learning Perspectives of Agent-Based Models: Practical Applications to Economic Crises and Pandemics with Python, R, Netlogo and Julia. It was edited by researchers at Carnegie Mellon University and the University of Porto, and a programmer.

"In our book, the authors highlight ABM as a powerful tool for studying economics, especially in the context of financial crises and pandemics, where traditional models have proven inadequate," explains Anand Rao, professor of applied data science and AI at Carnegie Mellon's Heinz College, one of the book's editors.

The book focuses on the emerging properties of heterogeneous agents that cannot be deduced from the characteristics of individual agents. Authors include practical examples and applications with R, Python, Julia, and Netlogo. In addition, different learning approaches, including game theory and artificial intelligence, are compared, highlighting the advantages of each in modeling economic phenomena.

"We are living in a time marked by continual crises that directly affect the economy," explains Pedro Campos, associate professor of economics at the University of Porto, one of the book's editors. "For example, the COVID-19 pandemic represented one of the most significant challenges since World War II, and financial turmoil and continuous conflicts highlight the world's instability. In this context, it is crucial that we have better ways to study and understand what is happening."

Adds Joaquim Margarido, a programmer with expertise in IT and the book's third editor: "The fields of ABM and MAS model the way society is organized, so developing such systems could help us advance general understanding and guide decisions."

More information:
Summarized from the book, Machine Learning Perspectives of Agent-Based Models: Practical Applications to Economic Crises and Pandemics with Python, R, Netlogo and Julia, edited by Campos, P (University of Porto), Rao, A (Carnegie Mellon University), and Margarido, J (programmer). Copyright 2025 The Editors, under exclusive license to Springer Nature Switzerland AG. All rights reserved.

Provided by Carnegie Mellon University's Heinz College