KIMEL: A kernel incremental metalearning algorithm

Authors:
Chunguang Li;Yan Ye;Qiuyuan Miao;Hui-Liang Shen
Affiliations:
Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, People's Republic of China;Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, People's Republic of China;Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, People's Republic of China;Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, People's Republic of China
Venue:
Signal Processing
Year:
2013

Citing 19
Cited 0

Feature detection from local energy

Pattern Recognition Letters
An introduction to support Vector Machines: and other kernel-based learning methods

An introduction to support Vector Machines: and other kernel-based learning methods
Independent component analysis: algorithms and applications

Neural Networks
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
No-reference image quality assessment based on DCT domain statistics

Signal Processing
A natural image quality evaluation metric

Signal Processing
Geometric Mean for Subspace Selection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Image quality assessment based on multiscale geometric analysis

IEEE Transactions on Image Processing
Reduced-reference IQA in contourlet domain

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Kernel Adaptive Filtering: A Comprehensive Introduction

Kernel Adaptive Filtering: A Comprehensive Introduction
No-reference image quality assessment using structural activity

Signal Processing
A variable step size LMS algorithm

IEEE Transactions on Signal Processing
A robust variable step-size LMS-type algorithm: analysis andsimulations

IEEE Transactions on Signal Processing
The Kernel Least-Mean-Square Algorithm

IEEE Transactions on Signal Processing
Direct kernel biased discriminant analysis: a new content-based image retrieval relevance feedback algorithm

IEEE Transactions on Multimedia
The least mean fourth (LMF) adaptive algorithm and its family

IEEE Transactions on Information Theory
Image quality assessment: from error visibility to structural similarity

IEEE Transactions on Image Processing
Blind Image Quality Assessment Using a General Regression Neural Network

IEEE Transactions on Neural Networks
Non-Negative Patch Alignment Framework

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.08

Visualization

Abstract

The Kernel method is a powerful tool for extending an algorithm from linear to nonlinear case. Metalearning algorithm learns the base learning algorithm, thus to improve performance of the learning system. Usually, metalearning algorithms exhibit faster convergence rate and lower Mean-Square Error (MSE) than the corresponding base learning algorithms. In this paper, we present a kernelized metalearning algorithm, named KIMEL, which is a metalearning algorithm in the Reproducing Kernel Hilbert Space (RKHS). The convergence analyses of the KIMEL algorithm are performed in detail. To demonstrate the effectiveness and advantage of the proposed algorithm, we firstly apply the algorithm to a simple example of nonlinear channel equalization. Then we focus on a more practical application in blind Image Quality Assessment (IQA). Experimental results show that the KIMEL algorithm has superior performance.