ON THE APPLICABILITY OF WIENER'S CANONICAL EXPANSIONS.
INFORMATION RESEARCH ASSOCIATES INC LEXINGTON MASS
Pagination or Media Count:
In previous papers, the application of Wieners Hermite-Laguerre expansion procedure to the multiple-alternative discrete-decision problem, with learning, characteristic of many pattern recognition problems was proposed. The applicability to problems in cybernetics, intelligence, and learning of the resulting Bayes type system was discussed. In the present paper the results of a subsequent analytical investigation which included a digital simulation of the procedure are summarized. Emphasis is on the aspects of realizability, convergence, and applicability of the method as regards the classes of stochastic inputs for which the procedure is valid. Author