On Kernelization of Supervised Mahalanobis Distance Learners
| dc.creator | Chatpatanasiri, Ratthachat | |
| dc.creator | Korsrilabutr, Teesid | |
| dc.creator | Tangchanachaianan, Pasakorn | |
| dc.creator | Kijsirikul, Boonserm | |
| dc.date | 2008-04-09 | |
| dc.date | 2009-01-30 | |
| dc.date.accessioned | 2026-07-07T12:35:19Z | |
| dc.date.available | 2026-07-07T12:35:19Z | |
| dc.description | This paper focuses on the problem of kernelizing an existing supervised Mahalanobis distance learner. The following features are included in the paper. Firstly, three popular learners, namely, "neighborhood component analysis", "large margin nearest neighbors" and "discriminant neighborhood embedding", which do not have kernel versions are kernelized in order to improve their classification performances. Secondly, an alternative kernelization framework called "KPCA trick" is presented. Implementing a learner in the new framework gains several advantages over the standard framework, e.g. no mathematical formulas and no reprogramming are required for a kernel implementation, the framework avoids troublesome problems such as singularity, etc. Thirdly, while the truths of representer theorems are just assumptions in previous papers related to ours, here, representer theorems are formally proven. The proofs validate both the kernel trick and the KPCA trick in the context of Mahalanobis distance learning. Fourthly, unlike previous works which always apply brute force methods to select a kernel, we investigate two approaches which can be efficiently adopted to construct an appropriate kernel for a given dataset. Finally, numerical results on various real-world datasets are presented. | |
| dc.description | 23 pages, 5 figures. There is a seriously wrong formula in derivation of a gradient formula of the "kernel NCA" in the two previous versions. In this new version, a new theoretical result is provided to properly account kernel NCA | |
| dc.identifier | https://arxiv.org/abs/0804.1441 | |
| dc.identifier | http://arxiv.org/abs/0804.1441 | |
| dc.identifier.uri | http://salesiana.dossiersoluciones.com/handle/123456789/217731 | |
| dc.subject | Machine Learning | |
| dc.subject | Artificial Intelligence | |
| dc.title | On Kernelization of Supervised Mahalanobis Distance Learners | |
| dc.type | text |