Efficient Output Kernel Learning for Multiple Tasks

11/18/2015
by   Pratik Jawanpuria, et al.
0

The paradigm of multi-task learning is that one can achieve better generalization by learning tasks jointly and thus exploiting the similarity between the tasks rather than learning them independently of each other. While previously the relationship between tasks had to be user-defined in the form of an output kernel, recent approaches jointly learn the tasks and the output kernel. As the output kernel is a positive semidefinite matrix, the resulting optimization problems are not scalable in the number of tasks as an eigendecomposition is required in each step. Using the theory of positive semidefinite kernels we show in this paper that for a certain class of regularizers on the output kernel, the constraint of being positive semidefinite can be dropped as it is automatically satisfied for the relaxed problem. This leads to an unconstrained dual problem which can be solved efficiently. Experiments on several multi-task and multi-class data sets illustrate the efficacy of our approach in terms of computational efficiency as well as generalization performance.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/22/2019

Multi-Task Kernel Null-Space for One-Class Classification

The one-class kernel spectral regression (OC-KSR), the regression-based ...
research
06/22/2022

Multi-task twin support vector machine with Universum data

Multi-task learning (MTL) has emerged as a promising topic of machine le...
research
02/21/2016

Multi-task and Lifelong Learning of Kernels

We consider a problem of learning kernels for use in SVM classification ...
research
06/06/2017

Classifying Documents within Multiple Hierarchical Datasets using Multi-Task Learning

Multi-task learning (MTL) is a supervised learning paradigm in which the...
research
11/10/2016

Multi-Task Multiple Kernel Relationship Learning

This paper presents a novel multitask multiple kernel learning framework...
research
06/27/2012

Bayesian Efficient Multiple Kernel Learning

Multiple kernel learning algorithms are proposed to combine kernels in o...
research
01/19/2023

Multi-output multilevel best linear unbiased estimators via semidefinite programming

Multifidelity forward uncertainty quantification (UQ) problems often inv...

Please sign up or login with your details

Forgot password? Click here to reset