Backpropagation neural networks are trained by adjusting initially random interconnecting weights according to the steepest local error surface gradient. The authors examine the practical implications of the arbitrary starting point on the error landscape of the ensuing trained network. The effects on network convergence and performance are tested empirically, varying parameters such as network size, training rate, transfer function and data representation. The data used are live process control data from an injection molding plant
A. E. Smith et al., "An Empirical Analysis of Backpropagation Error Surface Initiation for Injection Molding Process Control," Proceedings of the 1991 IEEE International Conference on Systems, Man, and Cybernetics, 1991. 'Decision Aiding for Complex Systems', Institute of Electrical and Electronics Engineers (IEEE), Jan 1991.
The definitive version is available at http://dx.doi.org/10.1109/ICSMC.1991.169905
1991 IEEE International Conference on Systems, Man, and Cybernetics, 1991. 'Decision Aiding for Complex Systems'
Engineering Management and Systems Engineering
Keywords and Phrases
Backpropagation Error Surface; Data Representation; Error Landscape; Injection Molding Process Control; Learning Systems; Network Convergence; Neural Nets; Neural Networks; Plastics Industry; Process Computer Control; Random Interconnecting Weights; Training Rate; Transfer Function
Article - Conference proceedings
© 1991 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.