Sciweavers

Share
ICANN
2011
Springer

Learning from Multiple Annotators with Gaussian Processes

8 years 6 months ago
Learning from Multiple Annotators with Gaussian Processes
Abstract. In many supervised learning tasks it can be costly or infeasible to obtain objective, reliable labels. We may, however, be able to obtain a large number of subjective, possibly noisy, labels from multiple annotators. Typically, annotators have different levels of expertise (i.e., novice, expert) and there is considerable diagreement among annotators. We present a Gaussian process (GP) approach to regression with multiple labels but no absolute gold standard. The GP framework provides a principled non-parametric framework that can automatically estimate the reliability of individual annotators from data without the need of prior knowledge. Experimental results show that the proposed GP multi-annotator model outperforms models that either average the training data or weigh individually learned single-annotator models.
Perry Groot, Adriana Birlutiu, Tom Heskes
Added 29 Aug 2011
Updated 29 Aug 2011
Type Journal
Year 2011
Where ICANN
Authors Perry Groot, Adriana Birlutiu, Tom Heskes
Comments (0)
books