類神經網路 (第3版 附光碟) | 誠品線上

類神經網路 (第3版 附光碟)

內容簡介

內容簡介 人類的頭腦約由1011 個神經元所組成，所有的訊息就在神經元與神經元間靠著軸突及樹突的發送與接收來傳遞。在這樣的一個過程中，所接收進來的各種訊息被分類或辨認，進而形成了人類的認知與思維。現在我們利用數學的計算來模擬神經元的運作，進而模擬神經網路的傳送，以期達到分類或辨認。類神經網路的特點為學習，學習的目的是要調整神經腱的大小，即調整加權係數，我們要探討各種就是學習法則。的類神經網路的模型及其加權係數的調整公式，也本書著重於利用類神經網路的方法於圖形辨識與最佳化問題之解決，因此將先介紹傳統的識別方法，再介紹類神經網路的各種理論及模型。本書提供基本的例子讓讀者容易了解，容易進入類神經網路的領域，在探討的多個模型中，均有自己提出的見解。

作者介紹

作者介紹 ■作者簡介黃國源

產品目錄

產品目錄 第一章 簡介1.1 圖型的定義與圖型識別的方法1.2 Decision-theoretic Approach的圖形識別與空間分割1.3 Pattern Recognition Systems1.4 Non-parametric & Parametric Methods1.5 人類頭腦的Neuron與模擬的Perceptron1.6 Two Class Data分佈的複雜性1.7 Activation Function1.8 Development History of Neural Networks1.9 Neural Network Applications第二章 DECISION-THEORETIC PATTERN RECOGNITION 決策理論的圖形識別Decision-theoretic Approach的圖形識別與Discriminant Functions2.2 Nonparametric Pattern Recognition非參數式之圖形識別：Using Discriminant Functions2.2.1 Linear discriminant functions for pattern recognition2.2.2 Nonlinear discriminant functions for pattern recognition2.2.3 Perpendicular bisector2.2.4 Minimum-distance classifier2.2.5 Minimum-distance classifier with respect to point sets (Piecewise-linear discriminant functions, Nearest-neighbor classification)2.2.6 N-nearest neighbor classification rule2.3 Parametric Pattern Recognition 參數式之圖形識別2.3.1 Bayes theorem (貝氏定理) and probability density function (pdf)2.3.2 Bayes (Parametric) classification rule (貝氏分類法則)2.3.3 Sequential classification2.3.4 Neyman-Pearson test2.3.5 Linear Classifier Design2.3.6 Feature selection2.3.7 Error estimation2.4 Unsupervised Pattern Recognition2.4.1 Minimum spanning tree (MST) clustering2.4.2 K-means clustering2.4.3 Hierarchical Clustering Using Dendrogram (Unsupervised Clustering) 2第三章 PERCEPTRON 認知器數學上解Decision Boundary之困難3.2 Perceptron3.3 Classification3.4 Training (Learning)3.5 Flowcharts of Perceptron3.6 Convergence Proof of Perceptron for Fixed Increment Training Procedure3.7 Perceptron for Logic Operation3.8 Layered Machine (Committee Machine Voting Machine)3.9 Multiclass Perceptrons3.10 Perceptron with Sigmoidal Activation Function and Learning byGradient Descent Method3.11 Modified Fixed-increment Training Procedure3.12 Multiclass Perceptron with Delta Learning Rule3.13 Widrow-Hoff Learning Rule3.14 Correlation Learning Rule第四章 MULTILAYER PERCEPTRON 多層認知器 Introduction4.2 設計Multilayer Perceptron with 1 Hidden Layer 解XOR的分類問題4.3 Gradient and Gradient Descent Method in Optimization4.4 Multilayer Perceptron (MLP) and Forward Computation4.5 Back-propagation Learning Rule (BP)4.5.1 Analysis4.5.2 Back-propagation learning algorithm of one-hidden layer perceptron (I)4.5.3 Back-propagation learning algorithm of one-hidden layer perceptron (II)4.6 Experiment of XOR Classification & Discussions4.7 On Hidden Nodes for Neural Nets4.8 Application - NETtalk：A Parallel Network That Learns to Read Aloud4.9 Functional-Link Net第五章 RADIAL BASIS FUNCTION NETWORK (RBF) 輻射基底函數網路 Introduction5.2 RBF Network 第一層的Learning Algorithm5.3 RBF Network 第二層的Learning Algorithm5.4 設計RBF Model to Classify XOR Patterns第六章 SUPPORT VECTOR MACHINE (SVM) 支持向量的分類器Introduction6.2 點到Hyperplane之距離6.3 Role of Support Vectors in Optimal Margin Classifier for LinearlySeparable Case6.4 Find Optimal Margin Classifier for Linearly Separable Case6.5 SVM for Nonseparable Patterns6.5.1 Primal Problem6.5.2 Dual Problem6.6 Feature Transformation and Support Vector Machine (SVM) – KernelSVM6.6.1 Primal Problem and Optimal Separating Hyperplane之建立6.6.2 在Dual Problem上求解新的Feature Space上的Support Vector Machine6.6.3 Gradient Ascent的調適性的方法求 Lagrange Multipliers6.7 Multiclss Classification Using Support Vector Machine6.7.1 Maximum Selection Classification System Using SVMs6.7.2 利用SVM 於數字辨識的樹狀分類系統 (Tree Classification System)6.7.3 Multi-class Classification Using Many Binary SVMs6.8 SVM Examples6.8.1 直接利用Lagrange method (沒有利用KKT conditions 的Lagrangemethod)6.8.2 利用加入KKT 的Lagrange method6.8.3 Support Vector Machine (SVM) Using Feature Transformation –Kernel SVM6.8 Exercise第七章 KOHONEN’S SELF-ORGANIZING NEURAL NET 自我組織的類神經網路 Winner-Take-All Learning Rule7.2 Kohonen’s Self-organizing Feature Maps7.3 Self-organizing Feature Maps於TSP第八章 PRINCIPAL COMPONENT NEURAL NET 主分量類神經網路Introduction8.2 Hebbian Learning Rule8.3 Oja的學習法則8.4 Neural Network of Generalized Hebbian Learning Rule8.5 Data Compression8.6 Effect of Adding One Extra Point along the Direction of ExistingEigenvector8.7 Neural network的PCA的應用第九章 HOPFIELD NEURAL NET9.1 Lyapunov Function9.2 Discrete Hopfield Model9.3 Analog Hopfield Model9.3.1 Circuits and Power9.3.2 Analog Hopfield Model9.4 Optimization Application of Hopfield Model to TSP9.5 與Hopfield Neural Net有關的研究與應用第十章 CELLULAR NEURAL NETWORK 蜂巢式類神經網路10.1 簡介10.2 蜂巢式類神經網路架構10.3 蜂巢式類神經網路的穩定性分析10.4 蜂巢式類神經網路與Hopfield神經網路的比較10.5 離散蜂巢式類神經網路第十一章 HAMMING NET11.1 Introduction11.2 Hamming Distance and Matching Score11.3 Hamming Net Algorithm11.4 Comparator第十二章 ADAPTIVE RESONANCE THEORY NET (ART)12.1 Introduction12.2 ART1 Neural Model12.3 Carpenter Grossberg ART1 Net的Algorithm12.4 Revised ART algorithm第十三章 FUZZY, CLUSTERING, AND NEURAL NETWORKS13.1 Fuzzy C-means Clustering Algorithm13.2 Fuzzy Perceptron13.3 Pocket Learning Algorithm13.4 Fuzzy Pocket參考文獻附錄Appendix A：Inner Product (內積)Appendix B：Line Property and Distance from Point to LineAppendix C：Covariance MatrixAppendix D：Gram–Schmidt Orthonormal ProcedureAppendix E：Lagrange Multipliers MethodAppendix F：Gradient, Gradient Descent and Ascent Methods in OptimizationAppendix G：Derivation of Oja’s learning ruleAppendix H：類神經網路程式實驗報告範例Appendix I：實驗報告範例之電腦程式Appendix J：MATLAB Program of PerceptronAppendix K：MATLAB Program of Multilayer PerceptronAppendix L：FORTRAN Program for PerceptronAppendix M：畫aX+bY+cZ+常數= 0的平面的Matlab電腦程式Appendix N：Support Vector Machine的數學推導

商品規格