Evolutionary Computation: Perspectives on Past and Future

Evolutionary computation is a sub-field of artificial intelligence and artificial life that uses biologically inspired methods to solve optimization problems, using iterative refinements of a set of solutions via change and selection. This approach, which began in the 1950s, constitutes a growing set of algorithms capable of solving a wide range of problems, divided into various types that differ in selection, mutation, and representation of candidate solutions. Its successful applications are counted in multiple domains, including, but not limited to, optimization, machine learning, robotics, and various areas that study living systems. Evolutionary computation has recently seen a revival, particularly in the study of open-ended evolution, with important implications for the future of AI. It has a unique potential to generate endless innovations and lead to a paradigm shift in the development of artificial intelligence and artificial intelligence life.