Multitask Learning Using Regularized Multiple Kernel Learning
Lists
Only you see this
Abstract
Empirical success of kernel-based learning algorithms is very much dependent on the kernel function used. Instead of using a single fixed kernel function, multiple kernel learning (MKL) algorithms learn a combination of different kernel functions in order to obtain a similarity measure that better matches the underlying problem. We study multitask learning (MKL) problems and formulate a novel MTL algorithm that trains coupled but nonidentical MKL models across the tasks. The proposed algorithm is especially useful for tasks that have different input and/or output space characteristics and is computationally very efficient. Empirical results on three data sets validate the generalization performance and the efficiency of our approach.
BibTeX
@inproceedings{gonen2011multitask,
author = {Mehmet Gönen and Melih Kandemir and Samuel Kaski},
title = {{Multitask Learning Using Regularized Multiple Kernel Learning}},
booktitle = {ICONIP},
year = {2011},
doi = {10.1007/978-3-642-24958-7\_58},
}Loading related content...