Characteristics of a deterministic supervised learning scheme

Author

Hu, Chia-lun J. ; Tan, Jeng Yoong

Author_Institution

Dept. of Electr. Eng., Southern Illinois Univ., Carbondale, IL, USA

fYear

1993

fDate

16-18 Aug 1993

Firstpage

107

Abstract

Supervised learning in a one-layered, hard-limited perceptron can be formulated into a set of linear inequalities containing the unknown weight coefficients. Solving these inequalities under a generalized separability condition by a noniterative method then achieves the goal of supervised learning. The speed of learning, the robustness of recognition, the automatic feature competition and automatic feature extraction, and other characteristics of this novel one-step learning system are then discussed in detail

Keywords

feature extraction; learning (artificial intelligence); perceptrons; automatic feature competition; automatic feature extraction; deterministic supervised learning scheme; generalized separability condition; linear inequalities; noniterative method; one-layered hard-limited perceptron; one-step learning system; recognition robustness; weight coefficients; Character recognition; Equations; Feature extraction; Linear matrix inequalities; Neural networks; Pattern recognition; Robustness; Supervised learning; Vectors; Virtual manufacturing;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1993., Proceedings of the 36th Midwest Symposium on

Conference_Location

Detroit, MI

Print_ISBN

0-7803-1760-2

Type

conf

DOI

10.1109/MWSCAS.1993.343054

Filename

343054