Asymmetric kernel in Gaussian Processes for learning target variance

被引：4

作者：

Pintea, S. L. ^{[1
]}

van Gemert, J. C. ^{[1
]}

Smeulders, A. W. M. ^{[2
]}

机构：

[1] Delft Univ Technol, Comp Vis Lab, Delft, Netherlands

[2] Univ Amsterdam, Intelligent Sensory Informat Syst, Amsterdam, Netherlands

来源：

PATTERN RECOGNITION LETTERS | 2018年 / 108卷

关键词：

Gaussian process; Kernel metric learning; Asymmetric kernel distances; Regression; MIXTURES; REGRESSION;

D O I：

10.1016/j.patrec.2018.02.026

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This work incorporates the multi-modality of the data distribution into a Gaussian Process regression model. We approach the problem from a discriminative perspective by learning, jointly over the training data, the target space variance in the neighborhood of a certain sample through metric learning. We start by using data centers rather than all training samples. Subsequently, each center selects an individualized kernel metric. This enables each center to adjust the kernel space in its vicinity in correspondence with the topology of the targets - a multi-modal approach. We additionally add descriptiveness by allowing each center to learn a precision matrix. We demonstrate empirically the reliability of the model. (C) 2018 Elsevier B.V. All rights reserved.

引用

页码：70 / 77

页数：8