Semi-supervised learning

>> Google Books

Semi-supervised learning / [edited by] Olivier Chapelle, Bernhard Schölkopf, Alexander Zien

資料種別:

電子ブック

出版情報:

Cambridge, Mass. ; London : MIT, c2006

形態:

1 online resource (x, 508 p.)

シリーズ名:

Adaptive computation and machine learning ;

著者名:

ISBN:

9780262255899 [0262255898] ((ebook) :)
9780262033589 [0262033585]

書誌ID:

EB00000098

フルテキスト

書誌詳細

注記:

Includes bibliographical references.
A comprehensive review of an area of machine learning that deals with the use of unlabeled data in classification problems, this text looks at state-of-the-art algorithms, applications benchmark experiments, and directions for future research.

主題:

Supervised learning (Machine learning)

目次情報:

Series Foreword

Preface

Introduction to Semi-Supervised Learning / 1：

Supervised, Unsupervised, and Semi-Supervised Learning / 1.1：

When Can Semi-Supervised Learning Work? / 1.2：

Classes of Algorithms and Organization of This Book / 1.3：

Generative Models / I：

A Taxonomy for Semi-Supervised Learning Methods / Matthias W. Seeger2：

The Semi-Supervised Learning Problem / 2.1：

Paradigms for Semi-Supervised Learning / 2.2：

Examples / 2.3：

Conclusions / 2.4：

Semi-Supervised Text Classification Using EM / N. C. Nigam ; Andrew McCallum ; Tom Mitchell3：

Introduction / 3.1：

A Generative Model for Text / 3.2：

Experminental Results with Basic EM / 3.3：

Using a More Expressive Generative Model / 3.4：

Overcoming the Challenges of Local Maxima / 3.5：

Conclusions and Summary / 3.6：

Risks of Semi-Supervised Learning / Fabio Cozman ; Ira Cohen4：

Do Unlabled Data Improve or Degrade Classification Performance? / 4.1：

Understanding Unlabeled Data: Asymptotic Bias / 4.2：

The Asymptotic Analysis of Generative Smei-Supervised Learning / 4.3：

The Value of Labeled and Unlabeled Data / 4.4：

Finite Sample Effects / 4.5：

Model Search and Robustness / 4.6：

Conclusion / 4.7：

Probabilistic Semi-Supervised Cluster with Constraints / Sugato Basu ; Mikhail Bilenko ; Arindam Banerjee ; Raymond J. Mooney5：

HMRF Model for Semi-Supervised Clustering / 5.1：

HMRF-KMeans Algorithm / 5.3：

Active Learning for Constraint Acquistion / 5.4：

Experimental Results / 5.5：

Related Work / 5.6：

Low-Density Separation / 5.7：

Transductive Support Vector Machines / Thorsten Joachims6：

Why Use Margin on the Test Set? / 6.1：

Experiments and Applications of the TSVMs / 6.4：

Solving the TSVM Optimization Problem / 6.5：

Connection to Related Approaches / 6.6：

Summary and Conclusions / 6.7：

Semi-Supervised Learning Using Semi-Definite Programming / Tijl De Bie ; Nello Cristianini7：

Relaxing SVM transduction / 7.1：

An Approximation for Speedup / 7.2：

General Semi-Supervised Learning Settings / 7.3：

Empirical Results / 7.4：

Summary and Outlook / 7.5：

Appendix

The Extended Schur Complement Lemma

Gaussian Processes and the Null-Category Noise Model / Neil D. Lawrence ; Michael I. Jordan8：

The Noise Model / 8.1：

Process Model and the Effect of the Null-Category / 8.3：

Posterior Inference and Prediction / 8.4：

Results / 8.5：

Discussion / 8.6：

Entropy Regularization / Yves Grandvalet ; Yoshua Bengio9：

Derivation of the Criterion / 9.1：

Optimization Algorithms / 9.3：

Related Methods / 9.4：

Experiments / 9.5：

Proof of Theorem 9.1 / 9.6：

Data-Dependent Regularization / Adrian Corduneanu ; Tommi S. Jaakkola10：

Information Regularization on Metric Spaces / 10.1：

Information Regularization and Relational Data / 10.3：

Graph-Based Models / 10.4：

Label Propogation and Quadratic Criterion / Olivier Delalleau ; Nicolas Le Roux11：

Label Propogation on a Similarity Graph / 11.1：

Quadratic Cost Criterion / 11.3：

From Transduction to Induction / 11.4：

Incorporating Class Prior Knowledge / 11.5：

Curse of Dimensionality for Semi-Supervised Learning / 11.6：

The Geometric Basis of Semi-Supervised Learning / Vikas Sindhwani ; Misha Belkin ; Partha Niyogi11.7：

Incorporating Geometry in Regularization / 12.1：

Algorithms / 12.3：

Data-Dependent Kernels for Semi-Supervised Learning / 12.4：

Linear Methods for Large-Scale Semi-Supervised Learning / 12.5：

Connections to Other Algorithms and Related Work / 12.6：

Future Directions / 12.7：

Discrete Regularization / Dengyong Zhou ; Bernhard Scholkopf13：

Discrete Analysis / 13.1：

Semi-Supervised Learning with Conditional Harmonic Mixing / Christopher J. C. Burges ; John C. Platt13.3：

Conditional Harmonic Mixing / 14.1：

Learning in CHM Models / 14.3：

Incorporating Prior Knowledge / 14.4：

Learning the Conditionals / 14.5：

Model Averaging / 14.6：

Change of Representation / 14.7：

Graph Kernels by Spectral Transforms / Xiaojin Zhu ; Jaz Kandola ; John Lafferty ; Zoubin Ghahramani15：

The Graph Laplacian / 15.1：

Kernels by Spectral Transforms / 15.2：

Kernel Alignment / 15.3：

Optimizing Alignment Using QCQP for Semi-Supervised Learning / 15.4：

Semi-Supervised Kernels with Order Restraints / 15.5：

Spectral Methods for Dimensionality Reduction / Lawrence K. Saul ; Kilian Weinberger ; Fei Sha ; Jihun Ham15.6：

Linear Methods / 16.1：

Graph-Based Methods / 16.3：

Kernel Methods / 16.4：

Modifying Distances / Alon Orlitsky ; Sajama16.5：

Estimating DBD Metrics / 17.1：

Computing DBD Metrics / 17.3：

Semi-Supervised Learning Using Density-Based Metrics / 17.4：

Conclusions and Future Work / 17.5：

Semi-Supervised Learning in Practice / V：

Large-Scale Algorithms / 18：

Cost Approximations / 18.1：

Subset Selection / 18.3：

Semi-Supervised Protein Classification Using Cluster Kernels / Jason Weston ; Christina Leslie ; Eugene Ie ; William Stafford Noble18.4：

Representation and Kernels for Protein Sequences / 19.1：

Semi-Supervised Kernels for Protein Sequences / 19.3：

Prediction of Protein Function from Networks / Hyunjung Shin ; Koji Tsuda19.4：

Graph-Based Semi-Supervised Learning / 20.1：

Combining Multiple Graphs / 20.3：

Experiments on Function Prediction of Proteins / 20.4：

Conclusion and Outlook / 20.5：

Analysis of Benchmarks / 21：

The Benchmark / 21.1：

Application of SSL Methods / 21.2：

Results and Discussion / 21.3：

Perspectives / VI：

An Augmented PAC Model for Semi-Supervised Learning / Maria-Florina Balcan ; Avrim Blum22：

A Formal Framework / 22.1：

Sample Complexity Results / 22.3：

Algorithmic Results / 22.4：

Related Models and Discussion / 22.5：

Metric-Based Approaches for Semi-Supervised Regression and Classification / Dale Schuurmans ; Finnegan Southey ; Dana Wilkinson ; Yuhong Guo23：

Metric Structure of Supervised Learning / 23.1：

Model Selection / 23.3：

Regularization / 23.4：

Classification / 23.5：

Transductive Inference and Semi-Supervised Learning / Vladimir Vapnik23.6：

Problem Settings / 24.1：

Problem of Generalization in Inductive and Transductive Inference / 24.2：

Structure of the VC Bounds and Transductive Inference / 24.3：

The Symmetrization Lemma and Transductive Inference / 24.4：

Bounds for Transductive Inference / 24.5：

The Structural Risk Minimization Principle for Induction and Transduction / 24.6：

Combinatorics in Transductive Inference / 24.7：

Measures of Size of Equivalence Classes / 24.8：

Algorithms for Inductive and Transductive SVMs / 24.9：

Semi-Supervised Learning / 24.10：

Conclusion: / 24.11：

Transductive Inference and the New Problems of Inference

Beyond Transduction: Selective Inference / 24.12：

A Discussion of Semi-Supervised Learning and Transduction / 25：

References

Notation and Symbols

Contributors

Index

Online Index

Series Foreword

Preface

Introduction to Semi-Supervised Learning / 1：

Supervised, Unsupervised, and Semi-Supervised Learning / 1.1：

When Can Semi-Supervised Learning Work? / 1.2：

Classes of Algorithms and Organization of This Book / 1.3：

続きを見る

東工大ブックレビュー

類似資料:

1 図書 The elements of statistical learning : data mining, inference, and prediction (: [hard]) Hastie, T. J, Tibshirani, R. J., Friedman, J. H. (Jerome H.) Springer	7 電子ブック Partially Supervised Learning Schwenker, Friedhelm Springer eBooks Computer Science, Springer Berlin Heidelberg
2 電子ブック Learning Theory and Kernel Machines Conference on Computational Learning Theory, Schölkopf, Bernhard, Warmuth, Manfred SpringerLink Books - AutoHoldings, Springer Berlin Heidelberg	8 電子ブック Partially Supervised Learning Hutchison, David, Kanade, Takeo Springer eBooks Computer Science, Springer Berlin Heidelberg
3 図書 Learning theory and Kernel machines : 16th Annual Conference on Learning Theory and 7th Kernel Workshop, COLT/Kernel 2003, Washington, DC, USA, August 24-27, … Workshop on Computational Learning Theory, Kernel Workshop, Schölkopf, Bernhard, Warmuth, Manfred Springer	9 電子ブック Partially Supervised Learning
4 図書 Kernel based algorithms for mining huge data sets : supervised, semi-supervised, and unsupervised learning Huang, Te-Ming, Kecman, V. (Vojislav), 1948-, Kopriva, Ivica Springer	10 電子ブック Partially Supervised Learning
5 図書 Boosting : foundations and algorithms (: pbk) Schapire, Robert E., Freund, Yoav The MIT Press	11 電子ブック Partially Supervised Learning Hutchison, David, Kanade, Takeo SpringerLink Books - AutoHoldings, Springer Berlin Heidelberg
6 電子ブック Handbook of Statistical Bioinformatics Henry Horng-Shing Lu, Bernhard Schölkopf, Hongyu Zhao, Lu, Henry Horng-Shing, Schölkopf, Bernhard SpringerLink Books - AutoHoldings, Springer Berlin Heidelberg	12 電子ブック Partially Supervised Learning Schwenker, Friedhelm SpringerLink Books - AutoHoldings, Springer Berlin Heidelberg