首页--工业技术论文--自动化技术、计算机技术论文--自动化基础理论论文--人工智能理论论文--人工神经网络与计算论文

基于函数连接型神经网络的极端学习机

学位论文数据集第4-5页
摘要第5-7页
ABSTRACT第7-9页
目录第10-13页
Contents第13-16页
CHAPTER 1 NEURAL NETWORK FUNDAMENTALS第16-32页
    1.1 Introduction第16-30页
        1.1.1 What is an artificial neural network第16-22页
            1.1.1.1 The Analogies of artiifcial neural network to the brain第16-18页
            1.1.1.2 Artificial Neural Network components第18-22页
        1.1.2 Neural Network History第22-24页
        1.1.3 Artiifcial neural network types第24-27页
            1.1.3.1 Single Layer Feedforward Networks (SLFNs)第24-25页
            1.1.3.2 Multilayer Feedforward Networks (MLFNs)第25-26页
            1.1.3.3 Recurrent Networks第26-27页
        1.1.4 The Training of Neural Networks第27-30页
            1.1.4.1 Supervised Training第27-29页
            1.1.4.2 Unsupervised Training第29-30页
                1.1.4.2.1 Clustering learning algorithms第29页
                1.1.4.2.2 Adaptive resonance theory and self-organizing map learning algorithms第29-30页
        1.1.5 Neural network Applications第30页
    1.2 Chapter summary第30页
    1.3 Thesis objective第30-31页
    1.4 The thesis organization第31-32页
CHAPTER 2 RADIAL BASIS FUNCTION AND FUNCTIONAL LINK NETWORKS第32-44页
    2.1 Radial basis function networks第32-35页
        2.1.1 Approximation properties of RBF networks第33-34页
        2.1.2 Compairson of RBF Networks and Multilayer Presptron (MPL)第34-35页
    2.2 Functional link network第35-42页
        2.2.1 Mathematical Essence of Functional link Neural Networks第36-38页
        2.2.2 The Radial Basis Functional link network第38-39页
        2.2.3 The RBFL Networks training第39-42页
            2.2.3.1 Learning RBFL Centers第40-42页
                2.2.3.1.1 Selecting RBF Centers Randomly from Training Sets第40页
                2.2.3.1.2 Selecting RBF Centers by Clustering第40-41页
                2.2.3.1.3 Supervised Learning of All Parameters第41-42页
    2.3 Chapter summary第42-44页
CHAPTER 3 ELM ALGORITHM第44-57页
    3.1 The ELM theory第44-49页
        3.1.1 Interpolation theorem第45-46页
        3.1.2 Universal approximation theorem第46-49页
    3.2 ELM algorithms第49-54页
        3.2.1 Universal approximation theorem第50-51页
        3.2.2 Universal approximation theorem第51-53页
        3.2.3 ELM for high order neural network (HONN)第53-54页
    3.3 Chapter summary第54-57页
CHAPTER 4 EXPERIMENTAL RESULTS第57-67页
    4.1 he implementation of the radial basis functional link network第57-58页
    4.2 Results and discussion第58-66页
        4.2.1 Benchmarking with a regression problem第59-61页
            4.2.1.1 Approximaiton of SinC' function第59-60页
            4.2.1.2 Approximation of Friedman Functions第60-61页
        4.2.2 Benchmarking with Real-World Function Approximation Problems第61-63页
            4.2.2.1 Abalone Age Prediction Application第61-62页
            4.2.2.2 California Housing Prediction Application第62-63页
        4.2.3 Benchmarking with Real-World classification Problems第63-64页
            4.2.3.1 Medical Diagnosis Application: Diabetes第63-64页
        4.2.4 HDPE process modeling第64-66页
            4.2.4.1 Modeling HDPE Process第64-66页
    4.3 Chapter summary第66-67页
CHAPTER 5 CONCLUSIONS AND FUTURE WORK第67-70页
    5.1 Conclusions第67页
    5.2 Future works第67-70页
REFRENECES第70-76页
APPENDICES第76-80页
    A.1 Description of HDPE Procss第76-77页
    A.2 Selection of vairables第77-78页
    A.3 Moorse-Penrose generalized in vers第78-80页
ACKNOWLEDGMENT第80-81页
RESEARCH ACHIEVEMENTS AND PUBLISHED PAPERS第81-83页
BREIF INTRODUCTION OF AUTHOR AND SUPERVISOR第83-84页
附件第84-85页

论文共85页,点击 下载论文
上一篇:水解反硝化工艺污水处理过程中污泥减量机理的研究
下一篇:基于GPGPU的一种回归测试预优化方法