Neural Networks for Pattern RecognitionThis book provides the first comprehensive treatment of feed-forward neural networks from the perspective of statistical pattern recognition. After introducing the basic concepts of pattern recognition, the book describes techniques for modelling probability density functions, and discusses the properties and relative merits of the multi-layer perceptron and radial basis function network models. It also motivates the use of various forms of error functions, and reviews the principal algorithms for error function minimization. As well as providing a detailed discussion of learning and generalization in neural networks, the book also covers the important topics of data processing, feature extraction, and prior knowledge. The book concludes with an extensive treatment of Bayesian techniques and their applications to neural networks. |
Contents
1 Statistical Pattern Recognition | 1 |
2 Probability Density Estimation | 33 |
3 SingleLayer Networks | 77 |
4 The Multilayer Perceptron | 116 |
5 Radial Basis Functions | 164 |
6 Error Functions | 194 |
7 Parameter Optimization Algorithms | 253 |
8 Preprocessing and Feature Extraction | 295 |
10 Bayesian Techniques | 385 |
Symmetric Matrices | 440 |
Gaussian Integrals | 444 |
Lagrange Multipliers | 448 |
Calculus of Variations | 451 |
Principal Components | 454 |
References | 457 |
477 | |
Other editions - View all
Common terms and phrases
activation function algorithm applications approach approximation back-propagation basis function networks Bayes Bayesian bias Chapter class C₁ classification problems coefficients component consider corresponding covariance matrix data points data set decision boundary denotes density estimation density function derivatives dimensionality discriminant function discussed in Section eigenvalues eigenvectors equations evaluate example expression feature feed-forward function y(x Gaussian give given gradient descent Hessian matrix hidden units input space input variables input vector kernel linear discriminant maximum likelihood minimization minimum mixture model multi-layer perceptron network mapping network outputs neural networks non-linear normal obtain optimal output units output variables parameters pattern recognition polynomial posterior distribution posterior probabilities pre-processing prior probabilities probability density procedure quadratic radial basis function regression regularization represents result search direction sigmoid simple solution sum-of-squares error function target data target values techniques term theorem tion training data training set transformation variance weight space weight vector