Predicting structured data

• Corporate Author: Neural Information Processing Systems Foundation
• Other Authors: Bakir, Gökhan
• Format: Electronic eBook
• Language: English
• Published: Cambridge, Mass. : MIT Press, ©2007.
• Series: Neural information processing series.
• Subjects: Machine learning.; Computer algorithms.; Kernel functions.; Data structures (Computer science); elektronické knihy; electronic books
• ISBN: 9780262255691
• Physical Description: 1 online zdroj (viii, 348 pages) : illustrations.

Table of Contents:

• Measuring Similarity with Kernels
• Discriminative Models
• Modeling Structure via Graphical Models
• Joint Kernel Maps / Jason Weston [and others]
• Support Vector Machine Learning for Interdependent and Structured Output Spaces / Yasemin Altun, Thomas Hofmann, and Ioannis Tsochandiridis
• Efficient Algorithms for Max-Margin Structured Classification / Juho Rousu [and others]
• Discriminative Learning of Prediction Suffix Trees with the Perceptron Algorithm / Ofer Dekel, Shai Shalev-Shwartz, and Yoram Singer
• A General Regression Framework for Learning String-to-String Mappings / Corinna Cortes, Mehryar Mohri, and Jason Weston
• Learning as Search Optimization / Hal Daume III and Daniel Marcu
• Energy-Based Models / Yann LeCun [and others]
• Generalization Bounds and Consistency for Structured Labeling / David McAllester
• Kernel Conditional Graphical Models / Fernando Perez-Cruz, Zoubin Ghahramani, and Massimiliano Pontil
• Density Estimation of Structured Outputs in Reproducing Kernel Hilbert Spaces / Yasemin Altun and Alex J. Smola
• Gaussian Process Belief Propagation / Matthias W. Seeger.