Home | Contact Info | Directory | About SCS | SCS New Buildings Info |   Giving to SCS   SCS Dean's Advisory Board
CALENDAR OF EVENTS

 

 SCS Calendar Events

 Search for Events by Date

 Submit an Event to the SCS Calendar

When: Monday, April 21, 12:00 p.m.- 1:30 p.m.

Where: NSH 1305 NULLvalue

Gideon Mann, Google

LTI Seminar

Abstract:
Current machine learning systems can be effective when they have sufficient training data. However, human annotation is costly and it is too expensive to have humans hand-annotate training data for all classification tasks of interest. This dilema has led to the appeal of semi-supervised learning algorithms, where a small amount of labeled data is augmented by a larger pool of unannotated data. In this talk, I show how generalized expectation (GE) criteria can be used for semi-supervised learning. Unlike traditional semi-supervised learning methods that use conventionally labeled instances as their supervised seed information, GE makes use of labeled features, where individual features are labeled with their correlation with output labels. Experiments with logistic regression and conditional random fields on natural language processing problems demonstrate that training with GE on labeled features, as opposed to traditional supervised training and alternative semi-supervised learning methods, can substantially reduce the amount of time it takes to train high performance models.

<< Back