Recent studies utilize multiple kernel learning to deal with incomplete-data problem. In this study, we introduce new methods that do not only complete multiple incomplete kernel matrices simultaneously, but also allow control of the flexibility of the model by parameterizing the model matrix. By imposing restrictions on the model covariance, overfitting of the data is avoided. A limitation of kernel matrix estimations done via optimization of an objective function is that the positive definiteness of the result is not guaranteed. In view of this limitation, our proposed methods employ the LogDet divergence, which ensures the positive definiteness of the resulting inferred kernel matrix. We empirically show that our proposed restricted covariance models, employed with LogDet divergence, yield significant improvements in the generalization performance of previous completion methods.
Rachelle RIVERO
Gunma University,University of the Philippines
Tsuyoshi KATO
Gunma University,Waseda University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Rachelle RIVERO, Tsuyoshi KATO, "Parametric Models for Mutual Kernel Matrix Completion" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 12, pp. 2976-2983, December 2018, doi: 10.1587/transinf.2018EDP7139.
Abstract: Recent studies utilize multiple kernel learning to deal with incomplete-data problem. In this study, we introduce new methods that do not only complete multiple incomplete kernel matrices simultaneously, but also allow control of the flexibility of the model by parameterizing the model matrix. By imposing restrictions on the model covariance, overfitting of the data is avoided. A limitation of kernel matrix estimations done via optimization of an objective function is that the positive definiteness of the result is not guaranteed. In view of this limitation, our proposed methods employ the LogDet divergence, which ensures the positive definiteness of the resulting inferred kernel matrix. We empirically show that our proposed restricted covariance models, employed with LogDet divergence, yield significant improvements in the generalization performance of previous completion methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7139/_p
Copy
@ARTICLE{e101-d_12_2976,
author={Rachelle RIVERO, Tsuyoshi KATO, },
journal={IEICE TRANSACTIONS on Information},
title={Parametric Models for Mutual Kernel Matrix Completion},
year={2018},
volume={E101-D},
number={12},
pages={2976-2983},
abstract={Recent studies utilize multiple kernel learning to deal with incomplete-data problem. In this study, we introduce new methods that do not only complete multiple incomplete kernel matrices simultaneously, but also allow control of the flexibility of the model by parameterizing the model matrix. By imposing restrictions on the model covariance, overfitting of the data is avoided. A limitation of kernel matrix estimations done via optimization of an objective function is that the positive definiteness of the result is not guaranteed. In view of this limitation, our proposed methods employ the LogDet divergence, which ensures the positive definiteness of the resulting inferred kernel matrix. We empirically show that our proposed restricted covariance models, employed with LogDet divergence, yield significant improvements in the generalization performance of previous completion methods.},
keywords={},
doi={10.1587/transinf.2018EDP7139},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - Parametric Models for Mutual Kernel Matrix Completion
T2 - IEICE TRANSACTIONS on Information
SP - 2976
EP - 2983
AU - Rachelle RIVERO
AU - Tsuyoshi KATO
PY - 2018
DO - 10.1587/transinf.2018EDP7139
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2018
AB - Recent studies utilize multiple kernel learning to deal with incomplete-data problem. In this study, we introduce new methods that do not only complete multiple incomplete kernel matrices simultaneously, but also allow control of the flexibility of the model by parameterizing the model matrix. By imposing restrictions on the model covariance, overfitting of the data is avoided. A limitation of kernel matrix estimations done via optimization of an objective function is that the positive definiteness of the result is not guaranteed. In view of this limitation, our proposed methods employ the LogDet divergence, which ensures the positive definiteness of the resulting inferred kernel matrix. We empirically show that our proposed restricted covariance models, employed with LogDet divergence, yield significant improvements in the generalization performance of previous completion methods.
ER -