### Abstract

This paper studies the general problem of learning kernels based on a polynomial combination of base kernels. We analyze this problem in the case of regression and the kernel ridge regression algorithm. We examine the corresponding learning kernel optimization problem, show how that minimax problem can be reduced to a simpler minimization problem, and prove that the global solution of this problem always lies on the boundary. We give a projection-based gradient descent algorithm for solving the optimization problem, shown empirically to converge in few iterations. Finally, we report the results of extensive experiments with this algorithm using several publicly available datasets demonstrating the effectiveness of our technique.

Original language | English (US) |
---|---|

Title of host publication | Advances in Neural Information Processing Systems 22 - Proceedings of the 2009 Conference |

Pages | 396-404 |

Number of pages | 9 |

State | Published - 2009 |

Event | 23rd Annual Conference on Neural Information Processing Systems, NIPS 2009 - Vancouver, BC, Canada Duration: Dec 7 2009 → Dec 10 2009 |

### Publication series

Name | Advances in Neural Information Processing Systems 22 - Proceedings of the 2009 Conference |
---|

### Other

Other | 23rd Annual Conference on Neural Information Processing Systems, NIPS 2009 |
---|---|

Country | Canada |

City | Vancouver, BC |

Period | 12/7/09 → 12/10/09 |

### ASJC Scopus subject areas

- Information Systems

## Fingerprint Dive into the research topics of 'Learning non-linear combinations of kernels'. Together they form a unique fingerprint.

## Cite this

*Advances in Neural Information Processing Systems 22 - Proceedings of the 2009 Conference*(pp. 396-404). (Advances in Neural Information Processing Systems 22 - Proceedings of the 2009 Conference).