The 200th anniversary of Charles Darwin's birth coincides with the emerging ability of the Darwinian computational technique of genetic programming to automatically design complex structures, including analog electrical circuits, optical lens systems, controllers, antennas, quantum computers, mechanical systems, etc. The lecture will describe an automated "What You Want Is What You Get" process for designing complex structures based on the principles of natural selection, sexual recombination, and developmental biology.
The design process starts with the random creation of a primordial ooze of thousands of randomly created structure-constructing programs. Each program in the population specifies the steps by which a fully developed structure is to be progressively developed from a very simple embryonic structure. Each fully developed structure is translated into a data structure for input to an appropriate computer simulator (or other measuring method). The structure is then evaluated to obtain its behavior and performance. This yields a fitness measure indicating how well the structure satisfies the user's requirements. The population of program trees is genetically bred over a series of many generations using genetic programming. Genetic programming is driven by a fitness measure and employs genetic operations such as Darwinian reproduction, sexual recombination (crossover), and occasional mutation to create offspring. This automated evolutionary process produces both the topology and the numerical values for each component of the desired structure.
Download slides for this presentation in PDF format.
Some references to Charles Darwin, his life, his work, and its impact on modern thought.
About the speaker:
John R. Koza was co-founder, chairman, and CEO of Scientific Games Inc. between 1973 and 1987 where he co-invented the rub-off instant lottery ticket used by state lotteries. He has published four books and over 250 papers on genetic programming. He has taught a course on genetic algorithms and genetic programming at Stanford University in various years since 1988. He is currently a consulting professor in the Department of Electrical Engineering at Stanford University. He is also currently Chair of National Popular Vote and author of Every Vote Equal: A State-Based Plan for Electing the President by National Popular Vote.
John R. Koza
Los Altos, CA 94023