Accession Number : AD0658527

Title :   RESEARCH ON SELF-ORGANIZING MACHINES.

Descriptive Note : Final rept.,

Corporate Author : STANFORD RESEARCH INST MENLO PARK CALIF

Personal Author(s) : Duda,Richard O.

Report Date : SEP 1967

Pagination or Media Count : 31

Abstract : This report reviews the results of a program of research on self-organizing machines. This research was primarily directed at establishing the conditions under which iterative machine-learning procedures can be guaranteed to converge. The most significant result of this effort was the development of a fairly complete theory of the linear machine, an important trainable multicategory pattern classifier. Theoretical results are summarized and related to the present state of research in pattern recognition. (Author)

Descriptors :   (*PATTERN RECOGNITION, MATHEMATICAL MODELS), (*LEARNING MACHINES, PATTERN RECOGNITION), ACCURACY, MAPPING(TRANSFORMATIONS), CONVERGENCE, ITERATIONS, TRAINING, CLASSIFICATION

Subject Categories : Information Science
      Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE