Self-Organizing Coalitions for Managing Complexity by Juan C. Burguillo introduces a framework for designing algorithms based on ideas from network theory, cellular automata, multi-agent systems, and game theory; each of these is used to find solutions to complex problems. Centrally, this is done by leveraging cooperation and communication to elicit self-organization. This book is appropriate for any researcher or graduate student who wishes to explore new ways of thinking about game theoretic problems, problems that lend themselves to population-based metaheuristics, multi-agent systems, or problems related to cellular automata. The majority of the algorithms and experiments in the book were implemented with the author’s own open-source CellNet software, which provides an easy route to begin investigating. The book is organized in a logical fashion, beginning with relevant background information in the first part. The second part demonstrates how the idea of self-organizing coalitions can improve the efficacy of various algorithms for tasks ranging from optimization to time series prediction. The final part of the book deals with evolutionary games, and nicely ties together the ideas and concepts from the previous two parts. The organization and behaviour of complex systems forms the root of the paradigm of algorithms presented in this book. This topic is studied across many disciplines from the social sciences to engineering to pure mathematics. Self-Organizing Coalitions for Managing Complexity undertakes the ambitious task of integrating various ideas from areas of study that are typically considered in isolation. The excellent high-level overview of selected ideas from network theory, cellular automata, multi-agent systems, and game theory in the first part of the book readies the reader to approach the second part. In the second part, the topics presented in the background section are deftly stitched together; it demonstrates how coalitions and neighbourhood structure

中文翻译：

---

Juan C. Burguillo：管理复杂性的自组织联盟

Juan C. Burguillo 的自组织联盟管理复杂性介绍了一个基于网络理论、元胞自动机、多智能体系统和博弈论思想的算法设计框架；每一个都用于寻找复杂问题的解决方案。总的来说，这是通过利用合作和交流来引发自组织来完成的。本书适合任何希望探索博弈论问题、适用于基于群体的元启发式的问题、多智能体系统或与细胞自动机相关的问题的新思维方式的研究人员或研究生。本书中的大部分算法和实验都是使用作者自己的开源 CellNet 软件实现的，这为开始研究提供了一条简单的途径。本书以合乎逻辑的方式组织，从第一部分的相关背景信息开始。第二部分展示了自组织联盟的想法如何提高各种算法对从优化到时间序列预测的任务的效率。本书的最后一部分涉及进化游戏，并将前两部分的思想和概念很好地联系在一起。复杂系统的组织和行为构成了本书中提出的算法范式的根源。该主题涉及从社会科学到工程再到纯数学的许多学科。管理复杂性的自组织联盟承担了一项雄心勃勃的任务，即整合通常被孤立考虑的研究领域的各种想法。本书第一部分对网络理论、元胞自动机、多智能体系统和博弈论的精选思想进行了出色的高级概述，使读者准备好进入第二部分。在第二部分，背景部分呈现的主题巧妙地拼接在一起；它展示了联盟和邻里结构如何