David E Goldberg
The Design of Innovating Machines
This project defines an effective computational basis for studying processes that are commonly called innovative and creative. Working from the principal investigator's groundbreaking studies of fast genetic algorithms (GAs)--search procedures based on the mechanics of natural selection and genetics--complete in 1993, this project gathers and integrates that argument into a single monograph; at the same time, it makes the crucial connection to innovation. The study is methodologically prescriptive and constructive, and viewed from one point perspective the study's value lies in its ability to develop algorithms that solve boundedly difficult combinatorial problems in subquadratic times, globally and reliably. Viewed from another perspective, the study constructively builds what amounts to a scientific, computational theory of innovation. Beyond this important contribution, this project maps out future directions in the areas of development and analogy and shows how the study of these topics is a natural outgrowth of the current framework.