I will present a personal view of what genetic algorithms are and why they work, pulling out a couple of highlights I've encountered that have influenced my opinions. I will show an animation of our local favorite GA -- CHC. I will also show some visualization tools developed in our group and a couple of research ideas that I think could profitably be pursued from here. I will close with a prediction for the future of GAs.

Biographical sketch:
J. David Schaffer is a Research Fellow at Philips Research where for the last 17 years he has lead a project on theory and applications of Evolutionary Computation (EC). For 10 of those years, he led the commercially successful development of EC-based optimizers for Philips SMD assembly equipment. He has also been involved in the development of prototype systems that automatically learn TV viewers' preferences.

He holds a BS in Aerospace Engineering, an MS in Systems Engineering, and a Ph.D. in Electrical Engineering. His PhD thesis introduced VEGA, the first multiobjective genetic algorithm. His main interests are in the area of evolving leaning machines, particularly neural networks.
His checkered background includes work as an airframe stress engineer, two years as a VISTA volunteer, research on municipal personnel systems, and managing biometrics for clinical research on psychotropic drugs. He is the author of some five dozen publications, and holds eleven US patents.
He was one of the founding members the International Society for Genetic Algorithms (now ISGEC) and is associate editor of the Evolutionary Computation Journal published by MIT Press, and serves on the advisory committee of the International Journal of Smart Engineering System Design.

Philips Electronics is a major multinational firm headquartered in the Netherlands and providing products in lighting, consumer electronics, semiconductors, and medical systems. Dr. Schaffer works at the Philips Research Labs in Briarcliff Manor, NY, USA.