Title:
FUZZY EXPERT SYSTEM FOR INTERPRETABLE RULE EXTRACTION FROM NEURAL NETWORKS
Document Type and Number:
WIPO Patent Application WO2001061647
Kind Code:
A3
Abstract:
A method and apparatus for extracting an interpretable, meaningful, and concise rules set from neural networks is presented. The method involves adjustement of gain parameter, lambda and the threshold, Tj for the sigmoid activation function of the interactive-or operator used in the extraction/development of a rule set from an artificial neural network. A multi-stage procedure involving coarse (500) and fine adjustment (514) is used in order to constrain the range of the antecedents of the extracted rules to the range of values of the inputs to the artificial neural network. Furthermore, the consequents of the extracted rules are provided based on degree of membership such that they are easily understandable by human beings. The method disclosed may be applied to any pattern recognition task, and is particularly useful in applications such as vehicle occupant sensing and recognition, object recognition, gesture recognition, and facial pattern recognition, among others.
Inventors:
SRINIVASA NARAYAN (US)
OWECHKO YURI (US)
OWECHKO YURI (US)
Application Number:
PCT/US2001/001629
Publication Date:
May 10, 2002
Filing Date:
January 18, 2001
Export Citation:
Assignee:
HRL LAB LLC (US)
SRINIVASA NARAYAN (US)
OWECHKO YURI (US)
SRINIVASA NARAYAN (US)
OWECHKO YURI (US)
International Classes:
G06K9/00; G06K9/66; G06N3/04; G06N7/04; (IPC1-7): G06N3/04
Foreign References:
| US5438644A | 1995-08-01 | |||
| US5299285A | 1994-03-29 |
Other References:
BENITEZ J M ET AL: "ARE ARTIFICIAL NEURAL NETWORKS BLACK BOXES?", IEEE TRANSACTIONS ON NEURAL NETWORKS, IEEE INC, NEW YORK, US, vol. 8, no. 5, 1 September 1997 (1997-09-01), pages 1156 - 1164, XP000699325, ISSN: 1045-9227
NARAZAKI H ET AL: "REORGANIZING KNOWLEDGE IN NEURAL NETWORKS: AN EXPLANATORY MECHANISM FOR NEURAL NETWORKS IN DATA CLASSIFICATION PROBLEMS", IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS. PART B: CYBERNETICS, IEEE SERVICE CENTER, US, vol. 26, no. 1, 1 February 1996 (1996-02-01), pages 107 - 117, XP000582905, ISSN: 1083-4419
NARAZAKI H ET AL: "REORGANIZING KNOWLEDGE IN NEURAL NETWORKS: AN EXPLANATORY MECHANISM FOR NEURAL NETWORKS IN DATA CLASSIFICATION PROBLEMS", IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS. PART B: CYBERNETICS, IEEE SERVICE CENTER, US, vol. 26, no. 1, 1 February 1996 (1996-02-01), pages 107 - 117, XP000582905, ISSN: 1083-4419
Download PDF:
Previous Patent: SILICON CHIP TOKEN AND METHODS FOR MAKING SAME
Next Patent: VISUAL ATTENTION LOCATION SYSTEM
Next Patent: VISUAL ATTENTION LOCATION SYSTEM