Abstract

A class of optical computing systems is introduced for solving symbolic logic problems that are characterized by a set of data objects and a set of relationships describing the data objects. The data objects and relationships are arranged into sets of facts and rules to form a knowledge base. The solutions to symbolic logic problems involve inferring conclusions to queries by applying logical inference to the facts and rules. The general structure of an inference machine is discussed in terms of rule-driven and query-driven control flows. As examples of a query-driven inference machine, two hybrid optical system architectures are presented which use matched-filter and mapped-template logic, respectively.

© 1986 Optical Society of America

Optical expert system based on matrix-algebraic formulation

Jack Y. Jau, F. Kiamilev, Y. Fainman, S. Esener, and Sing H. Lee
Appl. Opt. 27(24) 5170-5175 (1988)

Cellular logic architectures for optical computers

Toyohiko Yatagai
Appl. Opt. 25(10) 1571-1577 (1986)

Optical learning (inference) machines

George Eichmann and H. John Caulfield
Appl. Opt. 24(14) 2051-2054 (1985)

Optical production systems using neural networks and symbolic substitution

Elizabeth Botha, David Casasent, and Etienne Barnard
Appl. Opt. 27(24) 5185-5193 (1988)

Inference engine for expert system by using optical array logic

Masaya Iwata, Jun Tanida, and Yoshiki Ichioka
Appl. Opt. 31(26) 5604-5613 (1992)