Title:
ADAPTIVE SEQUENTIAL DETECTION NETWORK
Document Type and Number:
WIPO Patent Application WO2003085597
Kind Code:
A3
Abstract:
Sequential detection networks are provided that do not rely on statistical models for the source statistics such as source conditional density functions. Further, the present invention provides sequential detection networks that are adaptive to on-line changes in the source statistics and are thus applicable to the analysis of dynamic problems including those with complex density functions. The present invention also provides sequential detection networks that can automatically make a decision to either accept a next data sample or make a classification decision based upon cost determinations. Still further, the present invention provides sequential detection networks that can automatically make decisions on the order of sampling from a given set of data streams.
Inventors:
ERTIN EMRE (US)
PRIDDY KEVIN L (US)
PRIDDY KEVIN L (US)
Application Number:
PCT/US2003/009250
Publication Date:
September 10, 2004
Filing Date:
March 27, 2003
Export Citation:
Assignee:
BATTELLE MEMORIAL INSTITUTE (US)
ERTIN EMRE (US)
PRIDDY KEVIN L (US)
ERTIN EMRE (US)
PRIDDY KEVIN L (US)
International Classes:
G06K9/62; G06N3/04; (IPC1-7): G06F17/18
Other References:
CHENGAN GUO ET AL: "Temporal difference learning applied to sequential detection", IEEE TRANS. NEURAL NETW. (USA), IEEE TRANSACTIONS ON NEURAL NETWORKS, MARCH 1997, IEEE, USA, vol. 8, no. 2, March 1997 (1997-03-01), pages 278 - 287, XP002283525, ISSN: 1045-9227
CHENGAN GUO ET AL: "A learning sequential detection method based on neural networks", 1996, NEW YORK, NY, USA, IEEE, USA, 1996, pages 1409 - 1412 vol., XP002283526, ISBN: 0-7803-2912-0
JOUNY I ET AL: "M-ary sequential hypothesis tests for automatic target recognition", IEEE TRANS. AEROSP. ELECTRON. SYST. (USA), IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS, APRIL 1992, USA, vol. 28, no. 2, April 1992 (1992-04-01), pages 473 - 483, XP002283527, ISSN: 0018-9251
RUCK D W ET AL: "The multilayer perceptron as an approximation to a Bayes optimal discriminant function", IEEE TRANS. NEURAL NETW. (USA), IEEE TRANSACTIONS ON NEURAL NETWORKS, DEC. 1990, USA, vol. 1, no. 4, December 1990 (1990-12-01), pages 296 - 298, XP002283528, ISSN: 1045-9227
BERTSEKAS D P ET AL: "Neuro-dynamic programming: an overview", 1995, NEW YORK, NY, USA, IEEE, USA, 1995, pages 560 - 564 vol.1, XP002283529, ISBN: 0-7803-2685-7
V. GURALNIK ET AL: "Event Detection from Time Series Data", PROCEEDINGS OF THE 5TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, SAN DIEGO CALIFORNIA USA, 1999, pages 33 - 42, XP002283530
CHENGAN GUO ET AL: "A learning sequential detection method based on neural networks", 1996, NEW YORK, NY, USA, IEEE, USA, 1996, pages 1409 - 1412 vol., XP002283526, ISBN: 0-7803-2912-0
JOUNY I ET AL: "M-ary sequential hypothesis tests for automatic target recognition", IEEE TRANS. AEROSP. ELECTRON. SYST. (USA), IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS, APRIL 1992, USA, vol. 28, no. 2, April 1992 (1992-04-01), pages 473 - 483, XP002283527, ISSN: 0018-9251
RUCK D W ET AL: "The multilayer perceptron as an approximation to a Bayes optimal discriminant function", IEEE TRANS. NEURAL NETW. (USA), IEEE TRANSACTIONS ON NEURAL NETWORKS, DEC. 1990, USA, vol. 1, no. 4, December 1990 (1990-12-01), pages 296 - 298, XP002283528, ISSN: 1045-9227
BERTSEKAS D P ET AL: "Neuro-dynamic programming: an overview", 1995, NEW YORK, NY, USA, IEEE, USA, 1995, pages 560 - 564 vol.1, XP002283529, ISBN: 0-7803-2685-7
V. GURALNIK ET AL: "Event Detection from Time Series Data", PROCEEDINGS OF THE 5TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, SAN DIEGO CALIFORNIA USA, 1999, pages 33 - 42, XP002283530
Download PDF:
Previous Patent: COUNTING OF COAT HANGERS
Next Patent: MODULAR SYSTEM FOR PROCESSING AN IMAGE AND SOFTWARE KIT
Next Patent: MODULAR SYSTEM FOR PROCESSING AN IMAGE AND SOFTWARE KIT