Cognitively Inspired Neural Network for Situation Recognition
Interim rept. 15 Dec 2008-15 Dec 2009
AIR FORCE RESEARCH LAB HANSCOM AFB MA SENSORS DIRECTORATE/ELECTROMAGNETICS TECHNOLOGY DIV
Pagination or Media Count:
We developed a cognitively inspired mathematical learning framework called Neural Modeling Fields NMF. We applied it to learning and to recognition of situation composed of objects. NMF can successfully overcome the combinatorial complexity of associating subsets of objects with situations and provides fast and reliable convergence. We discuss the implications of the results of this current work for building multi layered intelligent systems.