Home Page of John R. Koza
John Koza with 1,000-Pentium parallel computer in Mountain View,
California. Picture by Eric Slomanson
Keynote speech at Congress On the Future of
Engineering Software (COFES) conference in
Most recent published article: Human-competitive results produced by genetic programming in March 2010 special 10th Anniversary issue of Genetic Programming and Evolvable Machines journal
Skip Down this Page
Directly to ...
Go to Biography
Go to Scientific books and papers
Go to Scientific research Interests
Go to Courses at Stanford University
Go to Recent Invited Talks and Tutorials on Genetic Programming
Go to Addresses and Contact Information
Go to Electoral College
Go to Miscellaneous
More pictures
Biography—John
R. Koza
Degrees
- B.A., Computer Science,
- M.A. Mathematics,
- M.S., Computer Science,
- PhD., Computer Science,
Stanford University
- Taught Computer Science
426 (Genetic Algorithms and Genetic Programming) at Stanford University in
Fall 1988, Spring 1989, Fall 1989, Spring 1990, Spring 1991,Spring 1992,
Spring 1993, Fall 1994, Fall 1995, Fall 1996, Winter 1997, Winter 1998;
Winter 1999, Winter 2000, Spring 2002, and Fall 2003 quarters. Click here
for information about MIS 226 /
CS 426 / EE392K course on genetic algorithms and genetic programming
at
- Taught Computer Science
425 (Artificial Life) at
- Co-taught, with Russ
Altman, Medical Information Science 214 in Spring 1995 (when the course
was first designed) and also co-taught this course in Spring 1996, Spring
l997, spring 1998, spring 1999, and spring 2000.
- Lecturer, Computer Science
Department, School of Engineering, 1988-1991.
- Consulting associate
professor, Computer Science Department, 1991–1992
- Consulting professor,
Computer Science Department,
- Consulting professor,
Stanford BioMedical Informatics, Department of
Medicine,
- Consulting professor,
Department of Electrical Engineering, School of Engineering, 1999 – April
2010.
- In the Business
Week survey of information technology by 800 international
researchers (June 12, 1997), Stanford placed first in two categories
(Computer Science and Artificial Intelligence) and second in
Telecommunications. In the remaining category in the survey
("BIOLOGICALLY INSPIRED: ARTIFICIAL LIFE, GENETIC ALGORITHMS"), the
Santa Fe Institute placed first, Stanford placed second, and Swiss Federal
Institute of Technology (EPFL, Lausanne) placed third. The ranking of SFI
and EPFL arises from the continuing active work of numerous individual
researchers.
Other Academic Activities
- Member of Board of
Trustees, Santa Fe Institute (1999
– 2002)
- Member of Science Board, Santa Fe Institute (1995 –
1998)
- Chair, 1996, 1997, and
1998 Genetic
Programming Conference
- Consulting editor, book series on
genetic programming with Kluwer Academic Publishers
- Member of Advisory Board
of the Genetic
Programming and Evolvable Machines journal published by Kluwer
Academic Publishers
- Member of editorial board
of Evolutionary Computation, Artificial Life, and Evolutionary
Optimization journals
- Associate editor of
IEEE Transactions on Evolutionary Computation (1997 – 2000)
- Member (since 1999) of
Executive Board of International Society for Genetic and Evolutionary
Computation (ISGEC)
- Member, Business Committee
of Genetic and Evolutionary Computation Conference (GECCO) for 1999 – 2005 and also
Program Policy Committee for Genetic Programming for 1999 – 2005
Industry
- Chairman, Chief Executive
Officer, and co-founder (1973-1987) of Scientific Games Inc
. of
20th anniversary party of Scientific Games Inc. in
- Co-inventor of the rub-off
instant lottery game ticket used by state lotteries since 1974
30th anniversary party held at the Massachusetts State
Lottery headquarters on May 24, 2004 celebrating the launch of the first
rub-off instant lottery game in May 1974. From left to right, Daniel W. Bower
(former president of Scientific Games Inc.), Joseph C. Sullivan (current
Executive Director of the Massachusetts State Lottery), Dr. William E. Perrault
(former Executive Director of the Massachusetts State Lottery in 1974 and
during the 1970s and early 1980s), and John R. Koza
(former Chairman and CEO of Scientific
Games Inc. from 1973 to 1987).
Photo courtesy of Public Gaming
magazine (July 2004 issue).
- President, Third
Millennium Venture Capital Limited (since l987), Third Millennium On-Line Products Inc.
(since 2000) and Genetic
Programming Inc. (since l998) which currently operates a 1,000-Pentium
parallel computer consisting of Pentium II 350 MHz microprocessors in a
Beowulf-style parallel cluster computer consisting of 533 MHz Alpha
microprocessors.
Electoral College
- Inventor 1966 board game Consensus concerning the electoral
college, featured on CBS Early Show TV on November 21, 2000
- Co-author of 2006 book
entitled Every Vote Equal: A State-Based
Plan For Electing the President By National Popular Vote and
originator of “Agreement
Among the States to Elect the President by National Popular Vote”
From interview on KGO TV on March 31, 2006 concerning “An Idea
To Make Your Vote Count In 2008”
Scientific
Books and Papers—John R. Koza
- 1992 book on genetic
programming entitled Genetic
Programming: On the Programming of Computers by Means of Natural Selection
from The MIT Press. The MIT Press also publishes a videotape entitled Genetic
Programming: The Movie associated with the first book. Click here for
more information about this 1992 videotape. The main points of
the 1992 book are
· Virtually all problems in artificial intelligence, machine learning, adaptive systems, and automated learning can be recast as a search for a computer program.
· Genetic programming provides a way to successfully conduct the search for a computer program in the space of computer programs.
- 1994 book on genetic
programming entitled Genetic
Programming II: Automatic Discovery of Reusable Programs from
The MIT Press.
The MIT Press also publishes a videotape entitled Genetic Programming II Videotape: The Next Generation. associated with this second book. Click here for additional information about this 1994 videotape. The main points of the 1994 book are
· Scalability is essential for solving non-trivial problems in artificial intelligence, machine learning, adaptive systems, and automated learning.
· Scalability can be achieved by reuse.
· Genetic programming provides a way to automatically discover and reuse subprograms in the course of automatically creating computer programs to solve problems.
- 1999 book Genetic
Programming III: Darwinian Invention and Problem Solving
from Morgan Kaufmann (by John R. Koza, Forrest H
Bennett III, David Andre, and Martin A. Keane). Morgan Kaufmann also
publishes Genetic Programming III Videotape: Human-Competitive Machine
Intelligence (by John R. Koza, Forrest H
Bennett III, David Andre, Martin A. Keane, and Scott Brave). Click here for
information about this 1999 videotape. The main points of the
1999 book is
· Genetic programming possesses the attributes that can reasonably be expected of a system for automatically creating computer programs.
- 2003 book Genetic
Programming IV: Routine
Human-Competitive Machine Intelligence from Kluwer Academic
Publishers (by John R. Koza, Martin A. Keane,
Matthew J. Streeter, William Mydlowec, Jessen
Yu, and Guido
· Genetic programming now routinely delivers high-return human-competitive machine intelligence.
· Genetic programming is an automated invention machine.
· Genetic programming can automatically create a general solution to a problem in the form of a parameterized topology.
· Genetic programming has delivered a progression of qualitatively more substantial results in synchrony with five approximately order-of-magnitude increases in the expenditure of computer time.
- Stanford University
technical reports from the Computer Science Department and Stanford BioMedical Informatics program in the Department of Medicine
of which I am author or co-author can be obtained on the web, including
- STAN-TR-CS
1314 (1990) entitled Genetic Programming: A Paradigm for
Genetically Breeding Populations of Computer Programs to Solve Problems
- STAN-TR-CS
1528 (1994) entitled Architecture-Altering Operations for
Evolving the Architecture of a Multi-Part Program in Genetic Programming
- STAN-TR-CS
1542 (1995) entitled Parallel Genetic Programming on a
Network of Transputers
- SMI-95-0586
(1995) entitled A Programming Course in Bioinformatics for Computer
and Information Science Students
- SMI-2000-0851
(2000) entitled Reverse Engineering and Automatic Synthesis of
Metabolic Pathways from Observed Data Using Genetic Programming
- Author (first or sole) on
about 200 papers and co-author of numerous other papers. Most of these
papers available PDF format.
- Click here for list of patents.
- Collaborators on various
published papers, books, or videos:
Al-Sakran, Sameer H.
Andre, David
Bade, Stephen L.
Comisky,
William
Dunlap, Frank
Fletcher, David
Hutchings, Jeffrey L.
Jones, Lee William
Lanza,
Guido
Lohn,
Jason
Poli,
Riccardo
Rice, James P.
Roughgarden,
Jonathan
Shipman, James
Tackett, Walter Alden
Yu, Jessen
Scientific
Research Interests—John R. Koza
Our main research interest is automatic programming (also called program
synthesis or program induction)—that is, getting computers to solve problems
without explicitly programming them.
This goal can be accomplished using the technique of genetic programming (of
which I am considered the inventor). Genetic programming is an automated method
for creating a working computer program from a high-level problem statement of
a problem. Genetic programming performs automatic program synthesis using
Darwinian natural selection and biologically inspired operations such as
recombination, mutation, inversion, gene duplication, and gene deletion. Old
Chinese saying says "animated gif is worth one megaword,"
so click
here for short tutorial of "What is GP?" For information
about the rapidly growing field of genetic programming, visit www.genetic-programming.org
and www.genetic-programming.com
While proof of principle ("toy") problems are occasionally useful for
tutorial or introductory purposes, we believe that it is time for fields of
artificial intelligence and machine learning to start delivering non-trivial
results that satisfy the test of being competitive with human performance.
Accordingly, our criterion for undertaking new research is that, if the
anticipated outcome of the research effort is achieved, it can be argued (on
some reasonable basis) that the result created by genetic programming is
competitive with human-produced results. Competitiveness with human performance
can be established in a variety of ways. For example, genetic programming may
produce a result that is slightly better, equal, or slightly worse than that
produced by a succession of human researchers working on an well-defined
problem over a period of years. Or, genetic programming may produce a result
that is equivalent to an invention that was patented in the past or that is
patentable today as a new invention. Or, genetic programming may produce a
result that is publishable in its own right (i.e., independent of the fact that
the result was mechanically generated). Or, genetic programming may produce a
result that wins or ranks highly in a judged competition involving human
contestants. There are examples using genetic programming in all four
categories and we have been produced at least one example in three of the four
categories. Fourteen are described in detail in the Genetic Programming III:
Darwinian Invention and Problem Solving book
and Human-Competitive Machine Intelligence videotape
For additional discussion, see human-competitive
machine intelligence.
Specifically, our recent research work involving genetic programming currently
emphasizes
- automated synthesis of
analog electrical circuits,
- automated synthesis of
controllers,
- automated synthesis
(reverse engineering) of metabolic pathways (networks of chemical
reactions),
- automated synthesis of
antennas,
- automated synthesis of
genetic networks,
- problems in computational
molecular biology,
- various other problems
involving cellular automata, multi-agent systems, mathematical algorithms,
and other areas of design, and
- using genetic programming
as an automated "invention machine" (for creating new and useful
patentable new inventions).
There are now a number of instances where genetic programming has
automatically produced a computer program that is competitive with human
performance. (See our criteria for human-competitive results and a list of human-competitive
results by clicking on human-competitive
machine intelligence). The fact that genetic programming can evolve
entities that are competitive with human-produced results suggests that genetic
programming may possibly be used as an "invention machine" to create
new and useful patentable inventions. In this connection, evolutionary methods,
such as genetic programming, have the advantage of not being encumbered by
preconceptions that limit human problem-solving to well-traveled paths.
In late July 1999, Genetic Programming Inc. started operating a new
1,000-node Beowulf-style parallel cluster computer consisting of 1,000 Pentium
II 350 MHz processors and a host computer. Genetic Programming Inc. has also
operated (starting in early 1999) a 70-node Beowulf-style parallel cluster
computer consisting of 533 MHz DEC Alpha microprocessors and a host computer.
The new 1,000-Pentium system is called the Tera-COTS
computer (since it has capacity of about a teraflops and is a beowulf-style customer computer made of "commodity
off-the-shelf" [COTS] parts). Click here for technical discussion of parallel genetic
programming and building the 1,000-Pentium
Beowulf-style parallel cluster computer.
All of the above-mentioned 21 human-competitive results were obtained
using computers that were substantially smaller than the new
1000-Pentium computer mentioned above. Fifteen of these 21 human-competitive
results were obtained on a 1995-vintage parallel computer system composed of 64
PowerPC 80 MHz processors with a spec95fp rating that is 1/60 of that of the
new 1000-Pentium machine. Five of these results were obtained on a 70-Alpha
machine (whose spec95fp rating is 1/9 of that of the 1000-Pentium machine). One
of these human competitive results were obtained with a 1994-vintage machine
(whose spec95fp rating is 1/1,320 of that of the 1000-Pentium machine). Because
of its increased computational power of the new 1000-Pentium machine, we expect
that it will produce additional human-competitive results.
Genetic programming has 16 important attributes that one would reasonably
expect of a system for automatic
programming (also called program synthesis or program
induction). Genetic programming has seven important
differences from other approaches to machine learning and artificial
intelligence.
My other research interests include artificial life (particularly
spontaneous emergence of self-replicating and self-improving entities) and
cellular automata.
Courses
at Stanford University
- Genetic Algorithms and
Genetic Programming Course: BMI 226 /
CS 426 / EE392K course on genetic algorithms and genetic programming is co-listed in the Department of Computer
Science in the School of Engineering, Department of Electrical Engineering
in the School of Engineering, and BioMedical
Informatics in the School of Medicine. Click here for additional
information about the course. This course is televised by
Stanford Instructional Television Network (SITN) operated by the Stanford Center for Professional
Development and is available on the WWW via Stanford On-Line.
Taught in Fall 1988, Spring 1989, Fall 1989, Spring 1990, Spring
1991,Spring 1992, Spring 1993, Fall 1994, Fall 1995, Spring 1997, Winter
1998, Winter 1998, Winter 1999, Winter 2000, Spring 2002, and Fall 2003
quarters. This course has been recorded on videotape starting in the
Spring 1997 quarter.
- Representations and
Algorithms for Computational Molecular Biology: This course is now
currently and regularly taught by Russ Altman,
M.D., Ph.D as BioMedical
Information Science 214 (BMI 214) in the School of Medicine (and co-listed
as Computer Science 274 in the School of Engineering). This course's home
page can be seen by clicking here: Course on
Computational Molecular Biology. This course was originally
taught by Russ Altman and John Koza, starting in
1995. Additional information about this course (primarily intended for
instructors of similar courses at other universities) was presented in a
paper by Russ Altman and John Koza at the 1996
Pacific Symposium on Biocomputing and in Stanford Medical Informatics
technical report SMI-95-0586
(1995) entitled A Programming Course in Bioinformatics for Computer and
Information Science Students. This course has been recorded on
videotape starting in the spring 1997 quarter, is televised, and is
available on the web via Stanford On-Line. For information, contact Stanford Center for Professional
Development. The Representations and Algorithms for
Computational Molecular Biology is part of a three-course
sequence at Stanford University that leads to a certificate in
Bioinformatics.
- Artificial Life Course -
Computer Science 425. This course was taught at
Compilations of the Student
Papers from 1994 to 2003 written by students in Computer Science 426
and by students in Computer Science 425 are available at the Mathematics
Library in the Main Quad at Stanford University and for purchase from the
Custom Publishing Department of the Stanford Bookstore.
Also, certain course
readers from John Koza's courses at
Stanford on Genetic Algorithms and Genetic Programming and course on Artificial
Life may be available from the Custom Publishing Department of the Stanford
Bookstore.
Recent Invited Talks and Tutorials—John R. Koza
·
Click here for PDF file of AAAI-2004 tutorial on automated
invention using genetic programming at American Association for
Artificial Intelligence conference (AAAI-2004) in
·
Click here for PDF file of
GECCO-2004 tutorial on genetic programming presented at the
Genetic and Evolutionary Computation Conference (GECCO)
in
·
Click here for PDF
file of EH-2004 invited talk on industrial-strength analog circuit synthesis by
means of genetic programming presented at the NASA/DoD
Conference on Evolvable Hardware (EH-2004) in
·
Click here for PDF
file of GPTP-2004 talk on industrial-strength analog circuit synthesis by means
of genetic programming presented at the Genetic Programming Theory and
Practice (GPTP) conference in
for the Study of Complex Systems.
John R. Koza E-Mail:
john at johnkoza
dot com · Click here for Awards for human-competitive results. · For information
about the field of genetic programming and the field of genetic and
evolutionary computation, visit www.genetic-programming.org
· The home page of
Genetic Programming Inc. at www.genetic-programming.com.
· The home page of
John R. Koza (including online versions of most published
papers) · 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. · 4,000+
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 Last updated September 19, 2010
Electoral
College
Addresses
and Contact Information—John R. Koza
Best
Address for Physical
Mail:
Post Office Box 1441
Los Altos Hills, California 94023-1411 USA
Miscellaneous