1st Edition

Manifold Learning Theory and Applications

Edited By Yunqian Ma, Yun Fu Copyright 2012
    330 Pages 128 B/W Illustrations
    by CRC Press

    Trained to extract actionable information from large volumes of high-dimensional data, engineers and scientists often have trouble isolating meaningful low-dimensional structures hidden in their high-dimensional observations. Manifold learning, a groundbreaking technique designed to tackle these issues of dimensionality reduction, finds widespread application in machine learning, neural networks, pattern recognition, image processing, and computer vision.

    Filling a void in the literature, Manifold Learning Theory and Applications incorporates state-of-the-art techniques in manifold learning with a solid theoretical and practical treatment of the subject. Comprehensive in its coverage, this pioneering work explores this novel modality from algorithm creation to successful implementation—offering examples of applications in medical, biometrics, multimedia, and computer vision. Emphasizing implementation, it highlights the various permutations of manifold learning in industry including manifold optimization, large scale manifold learning, semidefinite programming for embedding, manifold models for signal acquisition, compression and processing, and multi scale manifold.

    Beginning with an introduction to manifold learning theories and applications, the book includes discussions on the relevance to nonlinear dimensionality reduction, clustering, graph-based subspace learning, spectral learning and embedding, extensions, and multi-manifold modeling. It synergizes cross-domain knowledge for interdisciplinary instructions, offers a rich set of specialized topics contributed by expert professionals and researchers from a variety of fields. Finally, the book discusses specific algorithms and methodologies using case studies to apply manifold learning for real-world problems.

    Spectral Embedding Methods for Manifold Learning
    Introduction
    Spaces and Manifolds
    Data on Manifolds
    Linear Manifold Learning
    Nonlinear Manifold Learning
    Summary
    Acknowledgment

    Robust Laplacian Eigenmaps Using Global Information
    Introduction
    Graph Laplacian
    Global Information of Manifold
    Laplacian Eigenmaps with Global Information
    Experiments
    Summary
    Bibliographical and Historical Remarks

    Density Preserving Maps
    Introduction
    The Existence of Density Preserving Maps
    Density Estimation on Submanifolds
    Preserving the Estimated Density: The Optimization
    Summary
    Bibliographical and Historical Remarks

    Sample Complexity in Manifold Learning
    Introduction
    Sample Complexity of Classification on a Manifold
    Learning Smooth Class Boundaries
    Sample Complexity of Testing the Manifold Hypothesis
    Connections and Related Work
    Sample Complexity of Empirical Risk Minimization
    Relating Bounded Curvature to Covering Number
    Class of Manifolds with a Bounded Covering Number
    Fat-Shattering Dimension and Random Projections
    Minimax Lower Bounds on the Sample Complexity
    Algorithmic Implications
    Summary

    Manifold Alignment
    Introduction
    Formalization and Analysis
    Variants of Manifold Alignment
    Application Examples
    Summary
    Bibliographical and Historical Remarks
    Acknowledgments

    Large-scale Manifold Learning
    Introduction
    Background
    Comparison of Sampling Methods
    Large-Scale Manifold Learning
    Summary
    Bibliography and Historical Remarks

    Metric and Heat Kernel
    Introduction

    Theoretic Background
    Discrete Heat Kernel
    Heat Kernel Simplification
    Numerical Experiments
    Applications
    Summary
    Bibliographical and Historical Remarks

    Discrete Ricci Flow for Surface and 3-Manifold
    Introduction
    Theoretic Background
    Surface Ricci Flow
    3-Manifold Ricci Flow
    Applications
    Summary
    Bibliographical and Historical Remarks

    2D and 3D Objects Morphing Using Manifold Techniques
    Introduction
    Interpolation on Euclidean spaces
    Generalization of Interpolation Algorithms on a Manifold M
    Interpolation on SO(m)
    Application: The Motion of a Rigid Object in Space
    Interpolation on Shape Manifold
    Examples of Fitting Curves on Shape Manifolds
    Summary

    Learning Image Manifolds from Local Features
    Introduction
    Joint Feature-Spatial Embedding
    Solving the Out-Of-Sample Problem
    From Feature Embedding to Image Embedding
    Applications
    Summary
    Bibliographical and Historical remarks

    Human Motion Analysis Applications of Manifold Learning
    Introduction
    Learning A Simple Motion Manifold
    Factorized Generative Models
    Generalized Style Factorization
    Solving for Multiple Factors
    Examples
    Summary
    Bibliographical and Historical remarks

    Biography

    About the Editors:

    Yunqian Ma received his PhD in electrical engineering from the University of Minnesota at twin cities in 2003. He then joined Honeywell International Inc., where he is currently senior principal research scientist in the advanced technology lab at Honeywell Aerospace. He holds 12 U.S. patents and 38 patent applications. He has authored 50 publications, including 3 books. His research interest includes inertial navigation, integrated navigation, surveillance, signal and image processing, pattern recognition and computer vision, machine learning and neural networks. His research has been supported by internal funds and external contracts, such as AFRL, DARPA, HSARPA, and FAA. Dr. Ma received the International Neural Network Society (INNS) Young Investigator Award for outstanding contributions in the application of neural networks in 2006. He is currently associate editor of IEEE Transactions on Neural Networks, on the editorial board of the pattern recognition letters journal, and has served on the program committee of several international conferences. He also served on the panel of the National Science Foundation in the division of information and intelligent system and is a senior member of IEEE. Dr. Ma is included in Marquis Who is Who Engineering and Science.

    Yun Fu received his B.Eng. in information engineering and M.Eng. in pattern recognition and intelligence systems, both from Xian Jiaotong University, China. His M.S. in statistics, and Ph.D. in electrical and computer engineering, were both earned at the University of Illinois at Urbana-Champaign. He joined BBN Technologies, Cambridge, MA, as a Scientist in 2008 and was a part-time lecturer with the Department of Computer Science, Tufts University, Medford, MA, in 2009. Since 2010, he has been an assistant professor with the Department of Computer Science and Engineering, SUNY at Buffalo. His current research interests include applied machine learning, human-centered computing, pattern recognition, intelligent vision system, and social media analysis. Dr. Fu is the recipient of the 2002 Rockwell Automation Master of Science Award, Edison Cups of the 2002 GE Fund Edison Cup Technology Innovation Competition, the 2003 Hewlett-Packard Silver Medal and Science Scholarship, the 2007 Chinese Government Award for Outstanding Self-Financed Students Abroad, the 2007 DoCoMo USA Labs Innovative Paper Award (IEEE International Conference on Image Processing 2007 Best Paper Award), the 2007-2008 Beckman Graduate Fellowship, the 2008 M. E. Van Valkenburg Graduate Research Award, the ITESOFT Best Paper Award of 2010 IAPR International Conferences on the Frontiers of Handwriting Recognition (ICFHR), and the 2010 Google Faculty Research Award. He is a lifetime member of Institute of Mathematical Statistics (IMS), senior member of IEEE, member of ACM and SPIE.