Evolving Cellular Automata with Genetic Algorithms

Mitchell, M., Crutchfield, J., & Das, R. (1996). Evolving cellular automata with genetic algorithms: A review of recent work. Proceedings of the International Conference on Evolutionary Computing and Its Applications, 1, 1–14.

This paper investigates how evolution can create emergent computation, in which simple components that are only locally connected and informed can result in global information processing capabilities.

1D binary cellular automata (CAs) are used:

A CA is said to be performing a computation if some input can be encoded into an IC, and the CA transforms this IC into some output configuration which can be decoded into a meaningful result.

Density Classification

Evolution of Cellular Automata

Results

Extensions