Table of contents : Preface......Page 5 Acknowledgements......Page 7 Machines and brains......Page 8 The artificial neural network......Page 9 Introduction......Page 11 The performance of a single-neuron binary Perceptron......Page 14 Equivalent linear treshold function......Page 18 Learning a single-neuron binary Perceptron with the reinforcement rule......Page 20 The perceptron convergence theorem......Page 24 Performance of a two-layer binary Perceptron......Page 25 The adaptive recruitment learning rule......Page 30 Generalizing with a two-layer binary Perceptron......Page 32 The recruitment and reinforcement learning rule......Page 34 Application of the adaptive recruitment learning rule to switch circuits......Page 37 Application of the adaptive recruitment learning rule to hyphenation......Page 38 Application of the recruitment and reinforcement learning rule to contradictory binary data sets......Page 39 Intruduction......Page 41 The gradient descent adaptation method......Page 43 Learning with a single-neuron continous Perceptron......Page 46 The exacpt fitting of the data set with a single-neuron perceptron......Page 48 The approximate fiting of the data set with a single-neuron Perceptron......Page 50 Generalizing with a single-neuron continous Perceptron......Page 53 The classification of data with a single-neuron Perceptron......Page 54 Hyperplane boundary classification by one-zero labelling......Page 56 Hyperplane boundary classification by double treshold labelling......Page 61 Hyperplane boundary classification by single treshold labelling......Page 64 Application to the calssification of normally distributed classes......Page 68 Learning rule for a two-layer continuous Perceptron......Page 69 Under-fitting and over-fitting of a data set with a two-layer continuous Perceptron......Page 74 The class of functions realizable with a two-layer Perceptron......Page 78 The three-layer continuous Perceptron......Page 80 Application of a two-layer countinuous Perceptron to function indentification......Page 83 Application of a two-layer Perceptron to the mushroom classification problem......Page 84 Application of a two-layer Perceptron to the detection of the frequency of a sine wave......Page 85 Application of a multi-layer Perceptron to machine condition monitoring......Page 89 The learning speed of a continuous multi-layer Perceptron......Page 90 Initialization of weights and scaling the input and output......Page 91 Excercises......Page 92 Anthropomorphic pattern recognition with a self-organizing neural network......Page 93 The Bayes classification with a self-organinzing neural net algorithm......Page 98 Application of the self-organizing neural net algorithm to the classification of handwritten digits......Page 101 Topology preservation with a self-organizing algorithm......Page 103 Interpolation with self-organizing algorithm......Page 105 Master-slave and multi-net decomposition of the self-organizing neural net algorithm......Page 106 Application of the self-organizing algorithm to function identification......Page 107 Application of the self-organizing algorithm to robot arm control......Page 109 Application of the self-organizing algorithm to EEG signal analysis......Page 110 Application of the self-organizing algorithm to speech recognition......Page 112 Selecting and scaling of training vectors......Page 114 Some practical measures of performance of the self-organizing neural net algorithm......Page 115 Application of the self-organizing algorithm to signature identification......Page 118 Exercises......Page 119 Bibliography......Page 120 Index......Page 121