Evolving Cellular Automata: With Genetic Algorithms

本研究的目标是确定遗传算法在设计空间扩展、并行系统以执行计算任务方面的有用性。该项目的一个重要和新颖的组成部分是开发适当的理论框架，以理解由遗传算法进化的系统的计算策略以及设计这些策略的进化过程。本研究将做出三个主要贡献：(1)在空间扩展并行系统如何执行计算的研究方面取得进展；(2)遗传算法的理论与实践进展；(3)为图像处理、降噪、纹理检测和图像时间序列中相干结构的自动发现等并行计算问题的应用奠定基础。

The goals of this research are to determine the usefulness of genetic algorithms in designing spatially extended, parallel systems to perform computational tasks. An important and novel component of this project is the development of appropriate theoretical frameworks for understanding both the computational strategies of the systems evolved by the genetic algorithm as well as the evolutionary process by which such strategies are designed. This research will make three major contributions: (1) advances in the study of how spatially extended parallel systems perform computations; (2) advances in the theory and practice of genetic algorithms; and (3) a foundation for applications to parallel computation problems such as image processing, noise reduction, texture detection, and the automatic discovery of coherent structures in image time series.