Title :
Towards an information theoretic framework for object recognition
Author :
Westover, M. Brandon ; O´Sullivan, Joseph A.
Author_Institution :
Dept. of Phys., Washington Univ., St. Louis, MO, USA
fDate :
27 June-2 July 2004
Abstract :
We propose a model for rate-constrained pattern recognition problems, and present single-letter information bounds governing the conditions under which asymptotically error-free recognition is possible. The bounds depend on the statistics of the training and testing data, the number of pattern classes, and the rates of the codes used by the recognition system to internalize the data.
Keywords :
encoding; object recognition; problem solving; asymptotic error-free recognition; code rate; data testing; information theory; object recognition; pattern class; pattern testing; pattern training; rate-constrained pattern recognition; Electronic mail; Force sensors; Object recognition; Pattern recognition; Physics; Sensor systems; Statistical analysis; System testing; Systems engineering and theory; Training data;
Conference_Titel :
Information Theory, 2004. ISIT 2004. Proceedings. International Symposium on
Print_ISBN :
0-7803-8280-3
DOI :
10.1109/ISIT.2004.1365256
