Advanced architecture and training algorithms for recurrent neural networks
"Recurrent neural networks (RNN) attract considerable interest in computational intelligence fields because of their superior power in processing spatio-temporal data and time-varying signals. Traditionally, the recurrency of a neural network occurs between input samples along the time axis. The simultaneous recurrent network (SRN) extends the recurrent property to the spatial dimension. Presenting the feedback information with the same input vector to the network illustrates the transient properties of the system, which helps to trace error propagation and facilitates training. Backpropagation through time and extended Kalman filter are suitable gradient-based training algorithms for RNN"--Abstract, leaf iv.
Wunsch, Donald C.
Beetner, Daryl G.
Venayagamoorthy, Ganesh K.
Prokhorov, Danil V.
Dagli, Cihan H., 1949-
Vian, John L.
Sarangapani, Jagannathan, 1965-
Electrical and Computer Engineering
Ph. D. in Electrical Engineering
Boeing Company. Boeing Phantom Works
Mary K. Finley Missouri Endowment
National Science Foundation (U.S.)
University of Missouri--Rolla
Journal article titles appearing in thesis/dissertation
- Training winner-take-all simultaneous recurrent neural networks
- Evolutionary swarm neural network game engine for Capture Go
- Time series prediction with recurrent neural networks trained by a hybrid PSO-EA algorithm
x, 76 leaves
© 2006 Xindi Cai, All rights reserved.
Dissertation - Citation
Library of Congress Subject Headings
Back propagation (Artificial intelligence)
Evolutionary programming (Computer science)
Neural networks (Computer science)
Print OCLC #
Link to Catalog Record
Full-text not available: Request this publication directly from Missouri S&T Library or contact your local library.http://laurel.lso.missouri.edu/record=b5842377~S5
Cai, Xindi, "Advanced architecture and training algorithms for recurrent neural networks" (2006). Doctoral Dissertations. 1686.
This document is currently not available here.