Genetic Programming III: Darwinian Invention and Problem Solving - Hardcover

Koza, John R.; Bennett III, Forrest H.; Andre, David; Keane, Martin A.

Hardcover

ISBN 10: 1558605436 ISBN 13: 9781558605435

Publisher: Morgan Kaufmann, 1999

Synopsis

Genetic programming is a method for getting a computer to solve a problem by telling it what needs to be done instead of how to do it. Koza, Bennett, Andre, and Keane present genetically evolved solutions to dozens of problems of design, optimal control, classification, system identification, function learning, and computational molecular biology. Among the solutions are 14 results competitive with human-produced results, including 10 rediscoveries of previously patented inventions.

Researchers in artificial intelligence, machine learning, evolutionary computation, and genetic algorithms will find this an essential reference to the most recent and most important results in the rapidly growing field of genetic programming.

* Explains how the success of genetic programming arises from seven fundamental differences distinguishing it from conventional approaches to artificial intelligence and machine learning
* Describes how genetic programming uses architecture-altering operations to make on-the-fly decisions on whether to use subroutines, loops, recursions, and memory
* Demonstrates that genetic programming possesses 16 attributes that can reasonably be expected of a system for automatically creating computer programs
* Presents the general-purpose Genetic Programming Problem Solver
* Focuses on the previously unsolved problem of analog circuit synthesis, presenting genetically evolved filters, amplifiers, computational circuits, a robot controller circuit, source identification circuits, a temperature-measuring circuit, a voltage reference circuit, and more
* Introduces evolvable hardware in the form of field-programmable gate arrays
* Includes an introduction to genetic programming for the uninitiated

About this title

From the Back Cover

"Koza, Bennett, Andre, and Keane's evolutionary algorithm builds more complex and useful structures than the other approaches to computer learning that I have seen."
--John McCarthy, Stanford University

"John Koza and colleagues have demonstrated that genetic programming can be used to search highly discontinuous spaces and thereby find amazing solutions to practical engineering problems."
--Bernard Widrow, Stanford University

"In this impressive volume, the authors demonstrate that genetic programming is more than an intriguing idea-it is a practical synthesis method for solving hard problems."

-Nils J. Nilsson, Stanford University

"Through careful experiment, keen algorithmic intuition, and relentless application the authors deliver important results that rival those achieved by human designers. All readers in genetic and evolutionary computation and the related fields of artificial life, agents, and adaptive behavior will want this volume in their collections."

-David E. Goldberg, University of Illinois at Urbana-Champaign

"John Koza and his coauthors continue their relentless pursuit of a holy grail
in computer science: automatic programming."

-Moshe Sipper, Swiss Federal Institute of Technology (EPFL), Lausanne

Genetic programming is a method for getting a computer to solve a problem by telling it what needs to be done instead of how to do it. Koza, Bennett, Andre, and Keane present genetically evolved solutions to dozens of problems of design, optimal control, classification, system identification, function learning, and computational molecular biology. Among the solutions are 14 results competitive with human-produced results, including 10 rediscoveries of previously patented inventions.

Researchers in artificial intelligence, machine learning, evolutionary computation, and genetic algorithms will find this an essential reference to the most recent and most important results in the rapidly growing field of genetic programming.

Features

• Explains how the success of genetic programming arises from seven fundamental differences distinguishing it from conventional approaches to artificial intelligence and machine learning
  
• Describes how genetic programming uses architecture-altering operations to make on-the-fly decisions on whether to use subroutines, loops, recursions, and memory
  
• Demonstrates that genetic programming possesses 16 attributes that can reasonably be expected of a system for automatically creating computer programs
  
• Presents the general-purpose Genetic Programming Problem Solver
  
• Focuses on the previously unsolved problem of analog circuit synthesis, presenting genetically evolved filters, amplifiers, computational circuits, a robot controller circuit, source identification circuits, a temperature-measuring circuit, a voltage reference circuit, and more
  
• Introduces evolvable hardware in the form of field-programmable gate arrays
  
• Includes an introduction to genetic programming for the uninitiated

About the Author

John R. Koza is a consulting professor in the Section on Medical Informatics, Department of Medicine, School of Medicine at Stanford University. Forrest H Bennett III is chief scientist of Genetic Programming Inc., Los Altos, California. David Andre is a Ph.D. student in the Computer Science Division at the University of California at Berkeley.

Martin A. Keane is chief scientist of Econometrics, Inc., Chicago. Scott Brave is a research assistant in the Tangible Media Group at the MIT Media Laboratory.

Edition details

• Publisher: Morgan Kaufmann
• Publication date: 1999
• ISBN 10: 1558605436
• ISBN 13: 9781558605435
• Binding: Hardcover
• Language: English
• Edition number: 1
• Number of pages: 1154