Publication

Generalized Expectation Criteria for Semi-Supervised Learning with Weakly Labeled Data

Abstract
In this paper, we present an overview of generalized expectation criteria (GE) , a simple, robust, scalable method for semi-supervised training using weakly-labeled data. GE fits model parameters by favoring models that match certain expectation constraints, such as marginal label distributions, on the unlabeled data. This paper shows how to apply generalized expectation criteria to two classes of parametric models: maximum entropy models and conditional random fields. Experimental results demonstrate accuracy improvements over supervised training and a number of other state- of-the-art semi-supervised learning methods for these models.
Type
article
article
Date
2010-01-01
Publisher
Degree
Advisors
Rights
License
Research Projects
Organizational Units
Journal Issue
Embargo
Publisher Version
Embedded videos