This book constitutes the thoroughly refereed post-proceedings of the PASCAL (pattern analysis, statistical modelling and computational learning) Statistical and Optimization Perspectives Workshop on Subspace, Latent Structure and Feature Selection techniques, SLSFS 2005, held in Bohinj, Slovenia in February 2005.
The 9 revised full papers presented together with 5 invited papers were carefully selected during two rounds of reviewing and improvement for inclusion in the book. The papers reflect the key approaches that have been developed for subspace identification and feature selection using dimension reduction techniques, subspace methods, random projection methods, statistical analysis methods, Bayesian approaches to feature selection, latent structure analysis/probabilistic LSA, and optimisation methods.
Author(s): Wray Buntine, Aleks Jakulin (auth.), Craig Saunders, Marko Grobelnik, Steve Gunn, John Shawe-Taylor (eds.)
Series: Lecture Notes in Computer Science 3940 : Theoretical Computer Science and General Issues
Edition: 1
Publisher: Springer-Verlag Berlin Heidelberg
Year: 2006
Language: English
Pages: 209
Tags: Algorithm Analysis and Problem Complexity; Probability and Statistics in Computer Science; Computation by Abstract Devices; Artificial Intelligence (incl. Robotics); Image Processing and Computer Vision; Pattern Recognition
Front Matter....Pages -
Discrete Component Analysis....Pages 1-33
Overview and Recent Advances in Partial Least Squares....Pages 34-51
Random Projection, Margins, Kernels, and Feature-Selection....Pages 52-68
Some Aspects of Latent Structure Analysis....Pages 69-83
Feature Selection for Dimensionality Reduction....Pages 84-102
Auxiliary Variational Information Maximization for Dimensionality Reduction....Pages 103-114
Constructing Visual Models with a Latent Space Approach....Pages 115-126
Is Feature Selection Still Necessary?....Pages 127-138
Class-Specific Subspace Discriminant Analysis for High-Dimensional Data....Pages 139-150
Incorporating Constraints and Prior Knowledge into Factorization Algorithms – An Application to 3D Recovery....Pages 151-162
A Simple Feature Extraction for High Dimensional Image Representations....Pages 163-172
Identifying Feature Relevance Using a Random Forest....Pages 173-184
Generalization Bounds for Subspace Selection and Hyperbolic PCA....Pages 185-197
Less Biased Measurement of Feature Selection Benefits....Pages 198-208
Back Matter....Pages -
