preface
acknowledgements
abbreviations and symbols
glossary
introduction
��1 whatis aneuralnetwork?
��2 the human brain
��3 models of a neuron
��4 neural networks viewed as dirccted graphs
��5 feedback
6 network architecturns
7��knowledge representation
8 learning processes
9 learninglbks
��10 concluding remarks
��notes and rcferences
chapter 1 rosenblatt's perceptrou
��1.1 introduction
��1.2 perceptron
��1.3 1he pcrceptron convergence theorem
��1.4 relation between the perceptron and bayes classifier for a gaussian environment
��1.5 computer experiment��pattern classification
��1.6 the batch perceptron algorithm
��1.7 summary and discussion
notes and refercnces
problems
chapter 2 model building through regression
��2.1 introduction 68
2.2 linear regression model��preliminary considerafions
2.3 maximum a posteriori estimation ofthe parametervector
2.4 relationship between regularized least-squares estimation and map estimation
2.5 computer experiment��pattern classification
2.6 the minimum.description-length principle
2.7 rnite sample��size considerations
2.8 the instrumental��variables method
2 9 summary and discussion
notes and references
problems
chapter 3 the least��mean-square algorithm
��3.1 introduction
��3.2 filtering structure of the lms algorithm
��3.3 unconstrained optimization��a review
��3.4 thc wiener fiiter
��3.5 ne least.mean.square algorithm
��3.6 markov model portraying the deviation of the lms algorithm from the wiener filter
��3.7 the langevin equation��characterization ofbrownian motion
��3.8 kushner��s direct.averaging method
��3.9 statistical lms learning iheory for sinail learning��rate parameter
��3.10 computer experiment i:linear ptediction
��3.11 computer experiment ii��pattern classification
��3.12 virtucs and limitations of the lms aigorithm
��3.13 learning.rate annealing schedules
��3.14 summary and discussion
notes and refefences
problems
chapter 4 multilayer pereeptrons
��4.1 introductlon
��4.2 some preliminaries
��4.3 batch learning and on.line learning
��4.4 the back.propagation algorithm
��4 5 xorproblem
��4.6 heuristics for making the back��propagation algorithm perfoitn better
��4.7 computer experiment��pattern classification
��4.8 back propagation and differentiation
��4.9 the hessian and lis role 1n on-line learning
��4.10 optimal annealing and adaptive control of the learning rate
��4.11 generalization
��4.12 approximations of functions
��4.13 cross.vjlidation
��4.14 complexity regularization and network pruning
��4.15 virtues and limitations of back-propagation learning
��4.16 supervised learning viewed as an optimization problem
��4.17 couvolutionai networks
��4.18 nonlinear filtering
��4.19 small��seale versus large+scale learning problems
��4.20 summary and discussion
notes and rcfcreilces
problems
chapter 5 kernel methods and radial-basis function networks
��5.1 intreduction
��5.2 cover��s theorem on the separability of patterns
��5.3 1he interpolation problem
��5 4 radial��basis��function networks
��5.5 k.mcans clustering
��5.6 recursive least-squares estimation of the weight vector
��5 7 hybrid learning procedure for rbf networks
��5 8 computer experiment��pattern classification
��5.9 interpretations of the gaussian hidden units
��5.10 kernel regression and its relation to rbf networks
��5.11 summary and discussion
notes and references
problems
chapter 6 support vector machines
chapter 7 regularization theory
chapter 8 prindpal-components aaalysis
chapter 9 self-organizing maps
chapter 10 information-theoretic learning models
chapter 11 stochastic methods rooted in statistical mechanics
chapter 12 dynamic programming
chapter 13 neurodynami