Multiview Machine Learning (eBook)

Shiliang Sun, Liang Mao, Ziang Dong, Lidan Wu (Autoren)

eBook Download: PDF

2019 | 1st ed. 2019
X, 149 Seiten
Springer Singapore (Verlag)
978-981-13-3029-2 (ISBN)

Infos

This book provides a unique, in-depth discussion of multiview learning, one of the fastest developing branches in machine learning. Multiview Learning has been proved to have good theoretical underpinnings and great practical success. This book describes the models and algorithms of multiview learning in real data analysis. Incorporating multiple views to improve the generalization performance, multiview learning is also known as data fusion or data integration from multiple feature sets. This self-contained book is applicable for multi-modal learning research, and requires minimal prior knowledge of the basic concepts in the field. It is also a valuable reference resource for researchers working in the field of machine learning and also those in various application domains.  

Shiliang Sun received his Ph.D. degree in pattern recognition and intelligent systems from Tsinghua University, Beijing, China, in 2007. He is now a professor at the Department of Computer Science and Technology and the head of the Pattern Recognition and Machine Learning Research Group, East China Normal University, Shanghai, China. His current research interests include multiview learning, kernel methods, learning theory, probabilistic models, approximate inference, and sequential modeling. He has published 150+ research articles at peer-reviewed journals and international conferences. Prof. Sun is on the editorial board of several international journals, including IEEE Transactions on Neural Networks and Learning Systems, Information Fusion, and Pattern Recognition.

Liang Mao is a senior Ph.D. student at the Department of Computer Science and Technology and the Pattern Recognition and Machine Learning Research Group, East China Normal University, Shanghai, China. His main research interest is multiview learning and probabilistic models. 

Inhalt

Preface 5
Contents 6
1 Introduction 10
1.1 Background 10
1.2 Definition of Multiview Machine Learning and Related Concepts 10
1.3 Typical Application Fields in Artificial Intelligence 11
1.4 Why Can Multiview Learning Be Useful 13
1.5 Book Structure 14
References 15
2 Multiview Semi-supervised Learning 16
2.1 Introduction 16
2.2 Co-training Style Methods 17
2.2.1 Co-training 17
2.2.2 Co-EM 18
2.2.3 Robust Co-training 19
2.3 Co-regularization Style Methods 21
2.3.1 Co-regularization 21
2.3.2 Bayesian Co-training 23
2.3.3 Multiview Laplacian SVM 25
2.3.4 Multiview Laplacian Twin SVM 27
2.4 Other Methods 29
References 31
3 Multiview Subspace Learning 32
3.1 Introduction 32
3.2 Canonical Correlation Analysis and Related Methods 33
3.2.1 Canonical Correlation Analysis 33
3.2.2 Kernel Canonical Correlation Analysis 35
3.2.3 Probabilistic Canonical Correlation Analysis 37
3.2.4 Bayesian Canonical Correlation Analysis 38
3.3 Multiview Subspace Learning with Supervision 40
3.3.1 Multiview Linear Discriminant Analysis 40
3.3.2 Multiview Uncorrelated Linear Discriminant Analysis 42
3.3.3 Hierarchical Multiview Fisher Discriminant Analysis 44
3.4 Other Methods 45
References 46
4 Multiview Supervised Learning 47
4.1 Introduction 47
4.2 Multiview Large Margin Classifiers 48
4.2.1 SVM-2K 48
4.2.2 Multiview Maximum Entropy Discriminant 50
4.2.3 Soft Margin-Consistency-Based Multiview Maximum Entropy Discrimination 53
4.3 Multiple Kernel Learning 56
4.3.1 Kernel Combination 56
4.3.2 Linear Combination of Kernels and Support Kernel Machine 57
4.3.3 SimpleMKL 58
4.4 Multiview Probabilistic Models 60
4.4.1 Multiview Regularized Gaussian Processes 60
4.4.2 Sparse Multiview Gaussian Processes 61
4.5 Other Methods 63
References 64
5 Multiview Clustering 66
5.1 Introduction 66
5.2 Multiview Spectral Clustering 67
5.2.1 Co-trained Spectral Clustering 67
5.2.2 Co-regularized Spectral Clustering 68
5.3 Multiview Subspace Clustering 70
5.3.1 Multiview Clustering via Canonical Correlation Analysis 70
5.3.2 Multiview Subspace Clustering 71
5.3.3 Joint Nonnegative Matrix Factorization 73
5.4 Distributed Multiview Clustering 74
5.5 Multiview Clustering Ensemble 76
5.6 Other Methods 76
References 77
6 Multiview Active Learning 79
6.1 Introduction 79
6.2 Co-testing 80
6.3 Bayesian Co-training 81
6.4 Multiple-View Multiple-Learner 84
6.5 Active Learning with Extremely Spare Labeled Examples 86
6.6 Combining Active Learning with Semi-supervising Learning 88
6.7 Other Methods 90
References 90
7 Multiview Transfer Learning and Multitask Learning 91
7.1 Introduction 91
7.2 Multiview Transfer Learning with a Large Margin 92
7.3 Multiview Discriminant Transfer Learning 94
7.4 Multiview Transfer Learning with Adaboost 96
7.4.1 Adaboost 97
7.4.2 Multiview Transfer Learning with Adaboost 98
7.4.3 Multisource Transfer Learning with Multiview Adaboost 100
7.5 Multiview Multitask Learning 101
7.5.1 Graph-Based Interative Multiview Multitask Learning 101
7.5.2 Co-regularized Multiview Multitask Learning Algorithm 104
7.5.3 Convex Shared Structure Learning Algorithm for Multiview Multitask Learning 106
7.6 Other Methods 108
References 109
8 Multiview Deep Learning 111
8.1 Introduction 111
8.2 Joint Representation 112
8.2.1 Probabilistic Graphical Models 112
8.2.2 Fusion of Networks 116
8.2.3 Sequential Models 119
8.3 Complementary Structured Space 122
8.3.1 Deep Canonical Correlation Analysis 123
8.3.2 Methods Based on Autoencoders 125
8.3.3 Similarity Models 130
8.4 View Mapping 134
8.4.1 Generative Models 134
8.4.2 Retrieval-Based Methods 138
References 140
9 View Construction 145
9.1 Introduction 145
9.2 Feature Set Partition 146
9.2.1 Random Split 147
9.2.2 Genetic Algorithms 147
9.2.3 Pseudo Multiview Co-training 148
9.3 Purifying 148
9.4 Noising 150
9.5 Sequence Reversing 150
9.6 Multi-module 151
9.7 Conditional Generative Model 152
9.7.1 Conditional Generative Adversarial Nets 152
9.7.2 Conditional Variational Autoencoders 154
References 155

Produkt-Details

• Erscheint lt. Verlag: 7.1.2019
• Zusatzinfo: X, 149 p. 10 illus., 7 illus. in color.
• Verlagsort: Singapore
• Sprache: englisch
• Themenwelt: Informatik ► Datenbanken ► Data Warehouse / Data Mining
• Themenwelt: Informatik ► Grafik / Design ► Digitale Bildverarbeitung
• Themenwelt: Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik
• Schlagworte: Active learning • Clustering • data fusion • data integration • Deep learning • machine learning • multi-modal learning • multitask learning • multiview learning • Semi-Supervised Learning • subspace learning • supervised learning • transfer learning • View Construction
• ISBN-10: 981-13-3029-8 / 9811330298
• ISBN-13: 978-981-13-3029-2 / 9789811330292