Skip to content

Predicting Structured Data

Spend $50 to get a free DVD!

ISBN-10: 0262026171

ISBN-13: 9780262026178

Edition: 2007

Authors: Alexander J. Smola, Golchan Bakir, Thomas Hofmann, Bernhard Sch�lkopf, Ben Taskar

List price: $47.00
Blue ribbon 30 day, 100% satisfaction guarantee!
what's this?
Rush Rewards U
Members Receive:
Carrot Coin icon
XP icon
You have reached 400 XP and carrot coins. That is the daily max!

Description:

Machine learning develops intelligent computer systems that are able to generalize from previously seen examples. A new domain of machine learning, in which the prediction must satisfy the additional constraints found in structured data, poses one of machine learning's greatest challenges: learning functional dependencies between arbitrary input and output domains. This volume presents and analyzes the state of the art in machine learning algorithms and theory in this novel field. The contributors discuss applications as diverse as machine translation, document markup, computational biology, and information extraction, among others, providing a timely overview of an exciting field.…    
Customers also bought

Book details

List price: $47.00
Copyright year: 2007
Publisher: MIT Press
Publication date: 7/27/2007
Binding: Hardcover
Pages: 360
Size: 9.25" wide x 10.00" long x 0.75" tall
Weight: 1.980
Language: English

Alexander J. Smola is Senior Principal Researcher and Machine Learning Program Leader at National ICT Australia/Australian National University, Canberra.

Thomas Hofmann is a Director of Engineering at Google's Engineering Center in Zurich and Adjunct Associate Professor of Computer Science at Brown University.

Bernhard Sch�lkopf is Professor and Director at the Max Planck Institute for Biological Cybernetics in T�bingen, Germany. He is coauthor of Learning with Kernels (2002) and is a coeditor of Advances in Kernel Methods: Support Vector Learning (1998), Advances in Large-Margin Classifiers (2000), and Kernel Methods in Computational Biology (2004), all published by the MIT Press.

Ben Taskar is Assistant Professor in the Computer and Information Science Department at the University of Pennsylvania.

Preface
Introduction
Measuring Similarity with Kernels
Introduction
Kernels
Operating in Reproducing Kernel Hilbert Spaces
Kernels for Structured Data
An Example of a Structured Prediction Algorithm Using Kernels
Conclusion
Discriminative Models
Introduction
Online Large-Margin Algorithms
Support Vector Estimation
Margin-Based Loss Functions
Margins and Uniform Convergence Bounds
Conclusion
Modeling Structure via Graphical Models
Introduction
Conditional Independence
Markov Networks
Bayesian Networks
Inference Algorithms
Exponential Families
Probabilistic Context-Free Grammars
Structured Prediction
Conclusion
Structured Prediction Based on Discriminative Models
Joint Kernel Maps
Introduction
Incorporating Correlations into Linear Regression
Linear Maps and Kernel Methods
Generalizing Support Vector Machines
Joint Kernel Maps
Joint Kernels
Experiments
Conclusions
Support Vector Machine Learning for Interdependent and Structured Output Spaces
Introduction
A Framework for Structured/Interdependent Output Learning
A Maximum-Margin Formulation
Cutting-Plane Algorithm
Alternative Margin Formulations
Experiments
Conclusions
Proof of Proposition 37
Efficient Algorithms for Max-Margin Structured Classification
Introduction
Structured Classification Model
Efficient Optimization on the Marginal Dual Polytope
Experiments
Conclusion
Discriminative Learning of Prediction Suffix Trees with the Perceptron Algorithm
Introduction
Suffix Trees for Stream Prediction
PSTs as Separating Hyperplanes and the perceptron Algorithm
The Self-Bounded Perceptron for PST Learning
Conclusion
A General Regression Framework for Learning String-to-String Mappings
Introduction
General Formulation
Regression Problems and Algorithms
Pre-Image Solution for Strings
Speeding up Training
Comparison with Other Algorithms
Experiments
Conclusion
Learning as Search Optimization
Introduction
Previous Work
Search Optimization
Experiments
Summary and Discussion
Energy-Based Models
Introduction
Energy-Based Training
Architecture and Loss Function
Simple Architectures
Latent Variable Architectures
Analysis of Loss Functions for Energy-Based Models
Efficient Inference
Nonprobabilistic Factor Graphs
EBMs for Sequence Labeling and Structured Outputs
Conclusion
Generalization Bounds and Consistency for Structured Labeling
Introduction
PAC-Bayesian Generalization Bounds
Hinge Loss
Consistency
A Generalization of Theorem 62
Proofs of Theorems 61 and 62
Conclusion
Structured Prediction Using Probabilistic Models
Kernel Conditional Graphical Models
Introduction
A Unifying Review
Conditional Graphical Models
Experiments
Conclusions and Further Work
Density Estimation of Structured Outputs in Reproducing Kernel Hilbert Spaces
Introduction
Estimating Conditional Probability Distributions over Structured Outputs
A Sparse Greedy Optimization
Experiments: Sequence Labeling
Conclusion
Gaussian Process Belief Propagation
Introduction
Data and Model Dimension
Semiparametric Latent Factor Models
Gaussian Process Belief Propagation
Parameter Learning
Conclusions
References
Contributors
Index