Stochastic Low-Rank Kernel Learning for Regression
BibTeX
@MISC{Machart_stochasticlow-rank,
author = {Pierre Machart and Thomas Peel and Sandrine Anthoine and Liva Ralaivola and Hervé Glotin},
title = {Stochastic Low-Rank Kernel Learning for Regression},
year = {}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
We present a novel approach to learn a kernelbased regression function. It is based on the use of conical combinations of data-based parameterized kernels and on a new stochastic convex optimization procedure of which we establish convergence guarantees. The overall learning procedure has the nice properties that a) the learned conical combination is automatically designed to perform the regression task at hand and b) the updates implicated by the optimization procedure are quite inexpensive. In order to shed light on the appositeness of our learning strategy, we present empirical results from experiments conducted on various benchmark datasets. 1.
Citations
| 3341 | Pattern Classification and Scene Analysis - Duda, Hart - 1973 |
| 207 | Using the nyström method to speed up kernel machines - Williams, Seeger |
| 139 | Sparse greedy matrix approximation for machine learning - Smola, Schölkopf - 2000 |
| 60 | Input space versus feature space in kernelbased methods - Scholkopf, Mika, et al. - 1999 |
| 39 | A dual coordinate descent method for large-scale linear SVM - Hsieh, Chang, et al. - 2008 |
| 37 | Ridge regression: Applications to nonorthogonal problems. Technometrics 12 69– 82. [Correction: 12 723 - HOERL, W - 1970 |
| 9 | Efficiency of coordinate descent methods on huge-scale optimization problems - Nesterov |
| 5 | Ensemble nyström method - Kumar, Mohri, et al. - 2009 |
| 4 | Relationship of several variational methods for the approximate solution of ill-posed problems - Vasin - 1970 |
| 2 | Fast and accurate digit classification - Maji, Malik - 2009 |
| 1 | k A−1bb ⊤ A−1 − 1 k A−1b k b⊤ −1 1 A k ) , (30) Aronszajn, N. Theory of reproducing kernels - Drineas, Mahoney - 1950 |
