Normal view MARC view

Learning multiple tasks with kernel methods

Author: Evgeniou, Theodoros ; Michelli, Charles ; Pontil, MassimilianoINSEAD Area: Technology and Operations ManagementIn: Journal of Machine Learning Research, vol. 5, no. 4, April 2005 Description: p. 615-637.Type of document: INSEAD ArticleNote: Please ask us for this itemAbstract: The authors study the problem of learning many related tasks simultaneously using kernel methods and regularization. The standard single-task kernel methods, such as support vector machines and regularization networks, are extended to the case of multi-task learning. Their analysis shows that the problem of estimating many task functions with regularization can be cast as a single task learning problem if a family of multi-task kernel functions they define is used. These kernels model relations among the tasks and are derived from a novel form of regularizers. Specific kernels that can be used for multi-task learning are provided and experimentally tested on two real datasets. In agreement with past empirical work on multi-task learning, the experiments show that learning multiple related tasks simultaneously using the proposed approach can significantly outperform standard single-task learning particularly when there are many related tasks but few data per task.
Tags: No tags from this library for this title. Log in to add tags.
Item type Current location Call number Status Date due Barcode Item holds
INSEAD Article Europe Campus
Available BC007377
Total holds: 0

Ask Qualtrics

The authors study the problem of learning many related tasks simultaneously using kernel methods and regularization. The standard single-task kernel methods, such as support vector machines and regularization networks, are extended to the case of multi-task learning. Their analysis shows that the problem of estimating many task functions with regularization can be cast as a single task learning problem if a family of multi-task kernel functions they define is used. These kernels model relations among the tasks and are derived from a novel form of regularizers. Specific kernels that can be used for multi-task learning are provided and experimentally tested on two real datasets. In agreement with past empirical work on multi-task learning, the experiments show that learning multiple related tasks simultaneously using the proposed approach can significantly outperform standard single-task learning particularly when there are many related tasks but few data per task.

Digitized

There are no comments for this item.

Log in to your account to post a comment.
Koha 18.11 - INSEAD Catalogue
Home | Contact Us | What's Koha?