Now showing items 1-3 of 3

    • Intelligent System for Prediction of Sensor Drift 

      Golovko, V.; Savitsky, Y.; Sachenko, A.; Kochan, V.; Turchenko, V.; Laopoulos, T.; Grandinetti, L. (BPI, 1999)
      ln this paper the features of neural networks using for improve of measurement accuracy of physical quantities by sensor drift prediction are considered. There is use a technique of data volume increasing for training of predicting neural network by using of separate approximating neural network.


    • New Approach of the Recurrent Neural Network Training 

      Golovko, V.; Savitsky, Y. (BPI, 1999)
      In this work the technique o f creation o f adapthre training algorithms for recurrent neural networks (RNN) is cortsidered. These algorithms have high convergence and accuracy on a comparison with traditional backpropagation. The original technique of calculation of an adaptive training step with use ...


    • The Training of Feed-Forward Neural Networks 

      Golovko, V.; Dunets, A.; Savitsky, Y. (BPI, 1999)
      The training of multilayer perceptron is generally a difficult task. Excessive training times and lack of convergence to an acceptable solution are frequently reported. This paper discribes new training methods of feedforward neural networks. In comparison with standard backpropagation algorithm it ...