Gain Modification Enhances High Momentum Backward Propagation

Abstract

We present a backward propagation network which simultaneously modifies the gain parameters and the synaptic weights. Gain modification is shown to enhance the improvement in convergence rate obtained by high momentum in standard synaptic backward propagation. These improvements occur without degrading the generalization capabilities of the final solutions obtained by the network. Keywords: Neural nets; Backward propagation; Gain modification; Momentum; Effective time-dependent; Step constant.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 08, 1989
Accession Number
ADA216032

Entities

People

  • Charles M. Bachmann

Organizations

  • Brown University

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Convergence
  • Couplings
  • Differential Equations
  • Equations
  • Graphs
  • Military Research
  • Momentum
  • Neural Networks
  • Nonlinear Differential Equations
  • Notation
  • Rhode Island
  • Simulations
  • Standards
  • Training
  • United States
  • United States Government

Fields of Study

  • Physics

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Neural Network Machine Learning.
  • Ocean-Atmosphere Mesoscale Modeling, Data Assimilation, and Flux Boundary Layers