Parallel and Distributed Evolutionary Computation

This book provides a comprehensive and systematic introduction to the deep integration of evolutionary computation and parallel distributed computing. It analyzes how parallel and distributed architectures can overcome the bottlenecks and challenges faced by evolutionary algorithms in solving large-scale and distributed complex optimization problems, and constructs a new paradigm of distributed evolutionary computation that is controllable, scalable, and trustworthy. By combining the inherent parallelism of evolutionary computation with modern parallel and distributed computing environments, the book innovatively categorizes parallel distributed evolutionary computation into two core dimensions: (1) Parallel acceleration for centralized optimization, such as GPU acceleration and cluster-based parallelism; (2) Collaborative optimization for distributed scenarios, such as multi-agent cooperation. The book covers key technologies including population-level parallelism and dimension-level parallelism. It defines three major classes of optimization problems-dimension-distributed, data-distributed, and objective-distributed optimization-and provides in-depth discussions on cutting-edge challenges such as data privacy and security. It offers readers a comprehensive theoretical framework as well as practical guidance. This book can serve as a textbook for graduate students and senior undergraduates, and as a reference for researchers. It aims to help readers gain a deep understanding of the core principles, technical details, and application prospects of parallel distributed evolutionary computation, thereby supporting further research and innovation in evolutionary computation and distributed computing.