EVALUATION OF TWO METHODS FOR INSTRUCTING A CLASSIFIER,
Abstract
One of the problems in creating learning recognition systems is selecting a method of instruction which will assure maximum accuracy of recognition at the highest rate of convergence of the instruction process. Two methods of instructing a classifier are compared: the method used in the Adaline system and an instruction system in which the method of stochastic approximation is used. Accuracy of recognition and the convergence rate of the instruction process were selected as criteria for evaluating the quality of instruction. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 18, 1969
- Accession Number
- AD0700806
Entities
People
- L. A. Tyalksnis
- V. Yu. Chyarnyauskas
Organizations
- National Air and Space Intelligence Center