What people are saying - Write a review
We haven't found any reviews in the usual places.
Erkki Oja Helsinki University of Technology Finland
Josef Goppert Wolfgang Rosenstiel University of Tubingen Germany
Radiation Behavior of Analog Neural Network Chip
80 other sections not shown
adaptation application approach architecture artificial neural networks average backpropagation classification clusters components computed connection convergence criterion data set defined denotes density dimensional distribution epochs equation estimate example feedforward Figure function approximation Gaussian given gradient Hebbian Hebbian learning hidden neurons hidden nodes hidden units hyperplane IEEE IEEE Trans implementation initial input space input vector iteration kernel Kohonen kurtosis learning algorithm learning rate linear matrix mean square error method minimal multilayer multilayer perceptron neighborhood neuron noise nonlinear number of hidden obtained optimal output layer overfitting paper parameters perceptron performance prediction problem proposed quantization radial basis function random RBFN recurrent robust samples self-organizing map shown shows sigmoidal sigmoidal function signal simulation solution sparse square error stochastic structure supervised learning synapse target technique topology training algorithm training data training set University update variables vector quantization weight vector