Induction of Finite-State Languages Using Second-Order Recurrent Networks

Raymond L. Watrous; Gary M. Kuhn

doi:10.1162/neco.1992.4.3.406

More Information

Monthly

288 pp. per issue

6 x 9, illustrated

ISSN

0899-7667

E-ISSN

1530-888X

2014 Impact factor:

2.21

Neural Computation

May 1992, Vol. 4, No. 3, Pages 406-414

(doi: 10.1162/neco.1992.4.3.406)

Induction of Finite-State Languages Using Second-Order Recurrent Networks

Article PDF (447 KB)

Raymond L. Watrous

Siemens Corporate Research, 755 College Road East, Princeton, NJ 08540 USA

Gary M. Kuhn

Center for Communications Research, IDA, Thanet Road, Princeton, NJ 08540 USA

Abstract

Second-order recurrent networks that recognize simple finite state languages over {0,1}* are induced from positive and negative examples. Using the complete gradient of the recurrent network and sufficient training examples to constrain the definition of the language to be induced, solutions are obtained that correctly recognize strings of arbitrary length.