You are here

NEURAL NETWORKS FOR ADAPTIVE THREAT RESPONSE

Award Information
Agency: Department of Defense
Branch: Defense Advanced Research Projects Agency
Contract: N/A
Agency Tracking Number: 6090
Amount: $49,999.00
Phase: Phase I
Program: SBIR
Solicitation Topic Code: N/A
Solicitation Number: N/A
Timeline
Solicitation Year: N/A
Award Year: 1987
Award Start Date (Proposal Award Date): N/A
Award End Date (Contract End Date): N/A
Small Business Information
100 Allentown Pkwy - Ste 211
Allen, TX 75002
United States
DUNS: N/A
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 DR ROBERT L DAWES
 (214) 422-4570
Business Contact
Phone: () -
Research Institution
N/A
Abstract

NEURAL NETWORKS ARE A PROMISING SOLUTION TO SEVERAL PROBLEMS IN THE APPLICATION OF PARALLEL PROCESSING TECHNOLOGY FOR ADAPTIVE RECOGNITIONAND RESPONSE TO COMPLEX PATTERNS. UNFORTUNATELY, IT IS NOT KNOWN HOW TO GET A NEURAL NETWORK TO RECOGNIZE PATTERNS INDEPENDENTLY OF THE VALUES OF THEIR FEATURE PARAMETERS WITHOUT INCORPORATING SPECIAL INVARIANT TRANSFORMS FOR EACH SUCH FEATURE. THIS PROPOSAL INTRODUCES THE CONCEPT OF THE "PARAMETRIC AVALANCHE", WHICH DESCRIBES HOW A NEURAL NETWORK CAN ASSOCIATE PATTERNS THROUGH THE ACTIVATION OF FEATURE PATHWAYS OTHER THAN SIMPLE TIME PROXIMITY. THROUGH THIS MECHANISM, WE DESCRIBE NOT ONLY HOW TO MEASURE THE TRAJECTORY OF AN OBJECT, BUT ALSO HOW TO EXTRAPOLATE THE TRAJECTORY AND GUIDE A COUNTERMEASURE TO INTERCEPT THE TARGET.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government