1 February 1994 Recurrent back-propagation and Newton algorithms for training recurrent neural networks
Author Affiliations +
Proceedings Volume 2093, Substance Identification Analytics; (1994) https://doi.org/10.1117/12.172502
Event: Substance Identification Technologies, 1993, Innsbruck, Austria
Abstract
In this paper the recurrent back-propagation and Newton algorithms for an important class of recurrent networks and their convergence properties are discussed. To ensure proper convergence behavior, recurrent connections must be suitably constrained during the learning process. Simulation results demonstrate that the algorithms with the suggested constraint have superior performance.
© (1994) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Chung-Ming Kuan, Chung-Ming Kuan, Kurt Hornik, Kurt Hornik, Tung Liu, Tung Liu, } "Recurrent back-propagation and Newton algorithms for training recurrent neural networks", Proc. SPIE 2093, Substance Identification Analytics, (1 February 1994); doi: 10.1117/12.172502; https://doi.org/10.1117/12.172502
PROCEEDINGS
10 PAGES


SHARE
Back to Top