John Koza's Publications: Year Index:
Keane, Martin A., Koza, John R., and Streeter, Matthew J.
synthesis using genetic programming of an improved general-purpose controller
for industrially representative plants. In Stoica,
Adrian, Lohn, Jason, Katz, Rich, Keymeulen, Didier and Zebulum, Ricardo
(editors). Proceedings of 2002 NASA/DoD Conference
on Evolvable Hardware.
Most real-world controllers are composed of proportional, integrative, and derivative signal processing blocks. The so-called PID controller was invented and patented by Callender and Stevenson in 1939. In 1942, Ziegler and Nichols developed mathematical rules for automatically selecting the parameter values for PID controllers. In their influential 1995 book, Astrom and Hagglund developed a world-beating PID controller that outperforms the 1942 Ziegler-Nichols rules on an industrially representative set of plants. In this paper, we approached the problem of automatic synthesis of a controller using genetic programming without requiring in advance that the topology of the plant be the conventional PID topology. We present a genetically evolved controller that outperforms the automatic tuning rules developed by Astrom and Hagglund in 1995 for the industrially representative set of plants specified by Astrom and Hagglund.
Click here for a PDF version of this EH-2002 conference paper.
Keane, Martin A., Koza, John R., and Streeter, Matthew J.
2002a. Improved General-Purpose Controllers.
This patent application for three non-PID controllers and for PID tuning rules was filed on July 12, 2002. The controllers and tuning rules are described in the 2003 book Genetic Programming IV: Routine Human-Competitive Machine Intelligence.
Koza, John R. 2002a. Automatic synthesis
of both the topology and numerical parameters for complex structures using
genetic programming. In Chakrabarti, Amaresh (editor). Engineering Design
This chapter demonstrates that genetic programming can automatically create complex structures from a high-level statement of the structure's purpose. The chapter presents results produced by genetic programming that are from problem areas where there is no known general mathematical technique for automatically creating a satisfactory structure. The results include automatically synthesizing (designing) both the topology (graphical arrangement of components) and sizing (component values) for two illustrative analog electrical circuits and automatically synthesizing both the topology and tuning (component values) for a controller. Genetic programming not only succeeds in producing the required structure, but the structure is competitive with that produced by creative human designers. The claim that genetic programming has produced human-competitive results is supported by the fact that the automatically created results infringe on previously issued patents, improve on previously patented inventions, or duplicate the functionality of previously patented inventions.
Click here for a PDF version of this chapter from Chakrabarti’s edited book.
Koza, John R. (editor). 2002b. Genetic Algorithms and
Genetic Programming at Stanford 2002.
This volume contains 30 papers
written by students describing their term projects for the course "Genetic
Algorithms and Genetic Programming" (Medical Information Sciences 226 /
Computer Science 426) at
Click here for information on obtaining a copy of Book of Student Papers for 2002
Most of the student papers for 2002 are also available on-line.
Streeter, Matthew J., Keane, Martin A., and Koza, John R. 2002. Routine
duplication of post-2000 patented inventions by means of genetic programming.
In Foster, James A., Lutton, Evelyne, Miller, Julian, Ryan, Conor, and
Tettamanzi, Andrea G. B. (editors). 2002. Genetic Programming: 5th
European Conference, EuroGP 2002,
Previous work has demonstrated that genetic programming can automatically create analog electrical circuits, controllers, and other devices that duplicate the functionality and, in some cases, partially or completely duplicate the exact structure of inventions that were patented between 1917 and 1962. This paper reports on a project in which we browsed patents of analog circuits issued after January 1, 2000 on the premise that recently issued patents represent current research that is considered to be of practical and scientific importance. The paper describes how we used genetic programming to automatically create circuits that duplicate the functionality or structure of five post-2000 patented inventions. This work employed four new techniques (motivated by the theory of genetic algorithms and genetic programming) that we believe increased the efficiency of the runs. When an automated method duplicates a previously patented human-designed invention, it can be argued that the automated method satisfies a Patent-Office-based variation of the Turing test.
Click here for a PDF version of this Euro-GP-2002 conference paper.
Streeter, Matthew J., Keane, Martin A., and Koza, John R. 2002. Iterative
refinement of computational circuits using genetic programming. In Langdon, W.
B., Cantu-Paz, E., Mathias, K., Roy, R., Davis, D., Poli, R., Balakrishnan, K.,
Honavar, V., Rudolph, G., Wegener, J., Bull, L., Potter, M. A., Schultz, A. C.,
Miller, J. F., Burke, E., and Jonoska, N. (editors). Proceedings of the 2002
Genetic and Evolutionary Computation Conference.
Previous work has shown that genetic programming is capable of creating analog electrical circuits whose output equals common mathematical functions, merely by specifying the desired mathematical function that is to be produced. This paper extends this work by generating computational circuits whose output is an approximation to the error function associated with an existing computational circuit (created by means of genetic programming or some other method). The output of the evolved circuit can then be added to the output of the existing circuit to produce a circuit that computes the desired function with greater accuracy. This process can be performed iteratively. We present a set of results showing the effectiveness of this approach over multiple iterations for generating squaring, square root, and cubing computational circuits. We also perform iterative refinement on a recently patented cubic signal generator circuit, obtaining a refined circuit that is 7.2 times more accurate than the original patented circuit. The iterative refinement process described herein can be viewed as a method for using previous knowledge (i.e. the existing circuit) to obtain an improved result.
Click here for a PDF version of this GECCO-2002 conference paper.
· The home page of Genetic Programming Inc. at www.genetic-programming.com.
· For information about the field of genetic programming and the field of genetic and evolutionary computation, visit www.genetic-programming.org
· For information about John Koza’s course on genetic algorithms and genetic programming at Stanford University
· Information about the 1992 book Genetic Programming: On the Programming of Computers by Means of Natural Selection, the 1994 book Genetic Programming II: Automatic Discovery of Reusable Programs, the 1999 book Genetic Programming III: Darwinian Invention and Problem Solving, and the 2003 book Genetic Programming IV: Routine Human-Competitive Machine Intelligence. Click here to read chapter 1 of Genetic Programming IV book in PDF format.
· 3,440 published papers on genetic programming (as of November 28, 2003) in a searchable bibliography (with many on-line versions of papers) by over 880 authors maintained by William Langdon’s and Steven M. Gustafson.
· For information on the Genetic Programming and Evolvable Machines journal published by Kluwer Academic Publishers
· For information on the Genetic Programming book series from Kluwer Academic Publishers, see the Call For Book Proposals
information about the annual 2005
Genetic and Evolutionary Computation (GECCO) conference (which includes
the annual GP conference) to be held on June 25–29, 2005 (Saturday – Wednesday)
Last updated on August 23, 2004