Preface
Acknowledgements
AbbreviationsandSymbols
GLOSSARY
Introduction
1WhatisaNeuralNetwork?
2TheHumanBrain
3ModelsofaNeuron
4NeuralNetworksViewedAsDircctedGraphs
5Feedback
6NetworkArchitecturns
7 KnowledgeRepresentation
8LearningProcesses
9Learninglbks
10ConcludingRemarks
NotesandRcferences
Chapter1Rosenblatt'sPerceptrou
1.1Introduction
1.2Perceptron
1.31hePcrceptronConvergenceTheorem
1.4RelationBetweenthePerceptronandBayesClassifierforaGaussianEnvironment
1.5ComputerExperiment：PatternClassification
1.6TheBatchPerceptronAlgorithm
1.7SummaryandDiscussion
NotesandRefercnces
Problems
Chapter2ModelBuildingthroughRegression
2.1Introduction68
2.2LinearRegressionModel：PreliminaryConsiderafions
2.3MaximumaPosterioriEstimationoftheParameterVector
2.4RelationshipBetweenRegularizedLeast-SquaresEstimationandMAPEstimation
2.5ComputerExperiment：PatternClassification
2.6TheMinimum.Description-LengthPrinciple
2.7RniteSample—SizeConsiderations
2.8TheInstrumental，variablesMethod
29SummaryandDiscussion
NotesandReferences
Problems
Chapter3TheLeast—Mean-SquareAlgorithm
3.1Introduction
3.2FilteringStructureoftheLMSAlgorithm
3.3Unconstrainedoptimization：aReview
3.4ThCWienerFiIter
3.5neLeast.Mean.SquareAlgorithm
3.6MarkovModelPortrayingtheDeviationoftheLMSAlgorithmfromtheWienerFilter
3.7TheLangevinEquation：CharacterizationofBrownianMotion
3.8Kushner’SDirect.AveragingMethod
3.9StatisticalLMSLearningIheoryforSinailLearning—RateParameter
3.10ComputerExperimentI:LinearPTediction
3.11ComputerExperimentII：PatternClassification
3.12VirtucsandLimitationsoftheLMSAIgorithm
3.13Learning.RateAnnealingSchedules
3.14SummaryandDiscussion
NotesandRefefences
Problems
Chapter4MultilayerPereeptrons
4.1IntroductlOn
4.2SomePreliminaries
4.3BatchLearningandon.LineLearning
4.4TheBack.PropagationAlgorithm
45XORProblem
4.6HeuristicsforMakingtheBack—PropagationAlgorithmPerfoITnBetter
4.7ComputerExperiment：PatternClassification
4.8BackPropagationandDifferentiation
4.9TheHessianandlIsRole1nOn-LineLearning
4.10OptimalAnnealingandAdaptiveControloftheLearningRate
4.11Generalization
4.12ApproximationsofFunctions
4.13Cross.Vjlidation
4.14ComplexityRegularizationandNetworkPruning
4.15VirtuesandLimitationsofBack-PropagationLearning
4.16SupervisedLearningViewedasanOptimizationProblem
4.17COUVOlutionaINetworks
4.18NonlinearFiltering
4.19Small—SealeVerSusLarge+ScaleLearningProblems
4.20SummaryandDiscussion
NotesandRCfcreilces
Problems
Chapter5KernelMethodsandRadial-BasisFunctionNetworks
5.1Intreduction
5.2Cover’STheoremontheSeparabilityofPatterns
5.31heInterpolationProblem
54Radial—Basis—FunctionNetworks
5.5K.McansClustering
5.6RecursiveLeast-SquaresEstimationoftheWeightVector
57HybridLearningProcedureforRBFNetworks
58ComputerExperiment：PatternClassification
5.9InterpretationsoftheGaussianHiddenUnits
5.10KernelRegressionandItsRelationtoRBFNetworks
5.11SummaryandDiscussion
NotesandReferences
Problems
Chapter6SupportVectorMachines
Chapter7RegularizationTheory
Chapter8Prindpal-ComponentsAaalysis
Chapter9Self-OrganizingMaps
Chapter10Information-TheoreticLearningModels
Chapter11StochasticMethodsRootedinStatisticalMechanics
Chapter12DynamicProgramming
Chapter13Neurodynamics
Chapter14BayseianFilteringforStateEstimationofDynamicSystems
Chaptel15DynamlcaayDrivenRecarrentNetworks
Bibliography
Index
……
