MoL-2005-02: Leskes, Boaz (2005) The Value of Agreement: a new Boosting Algorithm. [Report]
Preview |
Text (Full Text)
MoL-2005-02.text.pdf Download (401kB) | Preview |
![[thumbnail of Abstract]](https://eprints.illc.uva.nl/style/images/fileicons/text.png) |
Text (Abstract)
MoL-2005-02.abstract.txt Download (1kB) |
Abstract
In the past few years unlabeled examples and their potential
advantage have received a lot of attention. In this paper a new
boosting algorithm is presented where unlabeled examples are used to
enforce agreement between several different learning algorithms. Not
only do the learning algorithms learn from the given training set but
they are supposed to do so while agreeing on the unlabeled
examples. Similar ideas have been proposed before (for example, the
Co-Training algorithm by Mitchel and Blum), but without a proof or
under strong assumptions. In our setting, it is only assumed that all
learning algorithms are equally adequate for the tasks. A new
generalization bound is presented where the use of unlabeled examples
results in a better ratio between training-set size and the the
resulting classifier's quality. The extent of this improvement
depends on the diversity of the learners--a more diverse group of
learners will result in a larger improvement whereas using two copies
of a single algorithm gives no advantage at all. As a proof of
concept, the algorithm, named AgreementBoost, is applied to two test
problems. In both cases, using AgreementBoost results in an up to 40%
reduction in the number of labeled examples.
| Item Type: | Report |
|---|---|
| Report Nr: | MoL-2005-02 |
| Series Name: | Master of Logic Thesis (MoL) Series |
| Year: | 2005 |
| Uncontrolled Keywords: | automated learning |
| Date Deposited: | 12 Oct 2016 14:38 |
| Last Modified: | 12 Oct 2016 14:38 |
| URI: | https://eprints.illc.uva.nl/id/eprint/756 |
Actions (login required)
 |
View Item |
