Kernel extrapolation

Authors:
S. V. N. Vishwanathan;Karsten M. Borgwardt;Omri Guttman;Alex Smola
Affiliations:
Statistical Machine Learning Program, National ICT Australia and RSISE, Australian National University, Canberra, 0200 ACT, Australia;Institute for Computer Science, Ludwig-Maximilians-University Munich, Oettingenstr. 67, 80538 Munich, Germany;Statistical Machine Learning Program, National ICT Australia and RSISE, Australian National University, Canberra, 0200 ACT, Australia;Statistical Machine Learning Program, National ICT Australia and RSISE, Australian National University, Canberra, 0200 ACT, Australia
Venue:
Neurocomputing
Year:
2006

Citing 5
Cited 3

Matrix analysis

Matrix analysis
Exploiting generative models in discriminative classifiers

Proceedings of the 1998 conference on Advances in neural information processing systems II
Diffusion Kernels on Graphs and Other Discrete Input Spaces

ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
Learning a kernel matrix for nonlinear dimensionality reduction

ICML '04 Proceedings of the twenty-first international conference on Machine learning
A kernel view of the dimensionality reduction of manifolds

ICML '04 Proceedings of the twenty-first international conference on Machine learning

Local linear transformation embedding

Neurocomputing
Learning an intrinsic-variable preserving manifold for dynamic visual tracking

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics - Special issue on game theory
Kernels for longitudinal data with variable sequence length and sampling intervals

Neural Computation

Quantified Score

Hi-index	0.01

Visualization

Abstract

We present a framework for efficient extrapolation of reduced rank approximations, graph kernels, and locally linear embeddings (LLE) to unseen data. We also present a principled method to combine many of these kernels and then extrapolate them. Central to our method is a theorem for matrix approximation, and an extension of the representer theorem to handle multiple joint regularization constraints. Experiments in protein classification demonstrate the feasibility of our approach.