Multi-module genetic programming with multiple genetic data representations

• US 7,424,461 B2
• Filed: 06/28/2001
• Issued: 09/09/2008
• Est. Priority Date: 06/28/2001
• Status: Active Grant

• Current Assignee: Valtrus Innovations Limited (f/k/a Dolya Holdco 9 Limited) (Key Patent Innovations Limited)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Kirshenbaum, Evan Randy
• Primary Examiner(s): Holmes; Michael B
• Application Number:
  US09/896,036
• Publication Number:
  US 20030004904A1
• Time in Patent Office: 2,630 Days
• Field of Search: 706 13- 44
• US Class Current: 706/13
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

A method for genetic programming which is based on multiple genetic data component representations. Multiple genetic data component representations are determined and a solution is evolved using a set of genetic data components represented by the multiple genetic data component representations.

Claims

1. A method for genetic programming, comprising the steps of:

determining a first genetic data component representation for expressing a solution model;

determining a second genetic data component representation for expressing the solution model;

generating a candidate solution by combining a genetic data component having the first genetic data component representation from a first parent solution with a genetic data component having the first genetic data component representation from a second parent solution and combining a genetic data component having the second genetic data component representation from the first parent solution with a genetic data component having the second genetic data component representation from the second parent solution; and

realizing the solution model as a recipe for constructing a structure that includes a neural network or an electrical circuit.

2. The method of claim 1, wherein the first and second genetic data component representations pertain to a first and a second module, respectively, of the solution model.

3. The method of claim 2, wherein the first and second modules include a program module.

4. The method of claim 3, wherein the genetic data component representation for the program module is a parse tree.

5. The method of claim 3, wherein the genetic data component representation for the program module is a sequence of instructions.

6. The method of claim 2, wherein the first and second modules include a parameters module.

7. The method of claim 6, wherein the genetic data component representation for the parameters module is a sequence of numbers.

8. The method of claim 7, wherein the sequence of numbers provides a set of parameters for a program module.

9. The method of claim 7, wherein the sequence of numbers provides a set of parameters for a math function module.

10. The method of claim 7, wherein the sequence of numbers provides a set of weights for a neural network module.

11. The method of claim 2, wherein the first and second modules include a neural network module.

12. The method of claim 11, wherein a the genetic data component representation of the neural network module is a parse tree.

13. The method of claim 2, wherein the first and second modules include a math function module.

14. The method of claim 13, wherein a the genetic data component representation of the math function module is a parse tree.

15. The method of claim 1, wherein the steps of combining include the step of matching the first and second genetic data component representations.